ADVANCES IN CLINICAL CHEMISTRY VOLUME 8
CONTRIBUTORS TO THIS VOLUME SHELDON F. G~TTLIEB G . I. C. INGRAM E. J.
VAN
...
14 downloads
1365 Views
19MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
ADVANCES IN CLINICAL CHEMISTRY VOLUME 8
CONTRIBUTORS TO THIS VOLUME SHELDON F. G~TTLIEB G . I. C. INGRAM E. J.
VAN
KAMPEN
ANDREW SASS-KORTSAK
DEREK WATSON
W. G. ZIJLSTRA
0
Advances in
CLINICAL CHEMISTRY Edited by
HARRY SOBOTKA Department of Chemistry, Mount Sinai Hospital N e w York, New York
C. P. STEWART Department of Clinical Chemistry, University of Edinburgh; Royal Infirmary, Edinburgh, Scotland
VOLUME 8
1965
ACADEMIC PRESS NEW Y O R K A N D LONDON
COPYRIGHT@ 1965, BY ACADEMIC PRESSINC. ALL RIQHTS RESERVED. NO PART OF THIS BOOK MAY BE REPRODUCED IN ANY FORM, BY PHOTOSTAT, MICROFILM, OR ANY OTHER MEANS, WITHOUT WRITEEN PERMISSION FROM THE PUBLISHERS.
ACADEMIC PRESS INC. 111 Fifth Avenue, New York, New York 10003
United Kingdom Edition published by ACADEMIC PRESS INC. (LONDON) LTD. Berkeley Square House, London W.1
LISRARY OF
CONGRESS C A T A m CARD
NUMBER: 58-12341
PRINTED IN THE UNITED STATES OF AMERICA
CONTRIBUTORS TO VOLUME 8 SHELDON F. GOTTLIEB, Department of Physiology, Department of Anesthesiology, Jefferson Medical College, Philadelphia, Pennsylvania G. I. C. INGRAM, Department of Haematology, Louis Jenner Laboratories, St. Thomas’ Hospital and Medical School, London, England
E. J. VAN KAMPEN,Clinical Chemical Laboratory, Diaconessenhuis, Groningen, The Netherlands ANDREW SASS-KORTSAK, Department of Paediatm’cs, University of Toronto, and The Research Institute, The Hospital for Sick Children, Toronto, Canada DEREKWATSON,~ Department of Biochemistry, Royal Women’s Hospital, Carlton, Victoria, Australia
W. G. ZIJLSTRA, Laboratory of Chemical Physiology, University of Groningen, Groningen, The Netherlands
‘Present address; Biochemistry Laboratories, Royal Columbian Hospital, New Wedminster, British Columbia, Canada. V
This Page Intentionally Left Blank
FOREWORD A historian of science in years to come may well be astonished a t the explosive burst of scientific activity round about the middle of the twentieth century of our era. He will be puzzled by the interrelationship between the growth of population and the rise of the standard of living; he will be interested in the increased percentage of scientists among the population, their greater specialization, and the resulting fragmentation of science; he will analyze the economic and the psychological motivation of scientists; he will compare the progress of knowledge with the broadness of the current of scientific publication. Living as we do in the midst of these events, we are hardly aware of their relatively rapid rate. What we notice is a doubling of the scientific output every ten years, regardless of contemporary political events. It is this climate which has engendered the appearance of series of reviews in dozens of disciplines. It may be with yearning or with a feeling of superiority that we look back a t such annual compendia as “Maly’s Jahresberichte der Thierchemie” of one hundred years ago, which encompassed the annual progress in the zoological half of biochemistry within 300 to 400 pages. Nowadays, that number of pages would not suffice to record the complete annual increment of knowledge in a single specialized division of a subject such as clinical chemistry. Media already existing furnish a comprehensive list of publications and an encyclopedic summarization of their contents; the present serial publication Advances in Clinical Chemistry-like other Advances-attempts something different. Its aim is to provide a readable account of selected important developments, of their roots in the allied fundamental disciplines, and of their impact upon the progress of medical science. The articles will be written by experts who are actually working in the field which they describe; they will be objectively critical discussions and not mere annotated bibliographies ; and the presentation of the subjects will be unbiased as the utterances of scientists are expected to be-sine ira et studio. The bibliography appended to each chapter will not only serve to document the author’s statements, it will lead the reader to those original publications in which techniques are described in full detail or in which viewpoints and opinions are expressed at greater length than is possible in the text. The selection of the subjects in the present and in future volumes will vii
viii
FOREWORD
include discussion of methods and of their rationale, critical and comparative evaluation of techniques, automation in Clinical Chemistry, and microanalytical procedures ; the contents will comprise those borderline subjects, such as blood coagulation or complement chemistry, which are becoming more chemical with increasing knowledge of the underlying reactions; in some instances the discussion of a subject will center around a metabolic mechanism or even around a disease entity. While recognizing that the elaboration and testing of methods is of the greatest importance in a subject, part of whose function is to provide reliable, accurate diagnostic and prognostic procedures, the new series will take cognizance of the fact that Clinical Chemistry plays an essential part in the progress of medical science in general by assisting in elucidating the fundamental biochemical abnormalities which underlie disease. The Editors hope that this program will stimulate the thinking of Clinical Chemists and of workers in related fields.
HARRY SOBOTKA C. P. STEWART
PREFACE This volume presents a cross section of clinical chemistry, and deals with mineral, gaseous, and protein constituents of blood and tissues. As in previous volumes, some articles emphasize analytical procedures whereas others deal with their topic from a general physiological and clinical viewpoint without neglecting the pertinent clinical chemistry laboratory methods. The editors’ concept of clinical chemistry, as stated in previous volumes of this serial publication, encompasses a wide field of subjects-from the mechanism of blood coagulation to the microbiological assay of vitamins, and from the significance of trace metals in health and disease to the therapeutic application of increased gas pressure. The natural history of clinical conditions and their comparison with normal controls share one common feature, namely, the central position of some species of molecules, be they small or large, all of them the natural subject of chemical research. Hence, practical methods develop, either manual or mechanized, the results of which become increasingly more useful for the clinician in diagnosis, therapy, and prognosis. As in the past, the editors wish to extend their thanks to the contributors and publishers.
October 1966
HARRYSOBOTKA C. P. STEWART
ix
This Page Intentionally Left Blank
CONTENTS . . . . . . . . . . . . . FOREWORD . . . . . . . . . . . . . . . . . PREFACE . . . . . . . . . . . . . . . . . LISTOF TABLES. . . . . . . . . . . . . . . . CONTRIBUTORS TO VOLUME 8
v
vii ix xiii
Copper Metabolism
ANDREW SASS-KORTSAK 1. 2 3. 4. 5.
Introduction and Historical Review ..................................... Methods of Investigation .............................................. Copper in Tissues and Biological Fluids ................................. Intake. Elimination. and Transport of Copper .......................... Physiological and Pathological Variations of Copper Metabolism ......... References ................................................................
.
1 3 15 24 32 50
Hyperbaric Oxygenation
SHELDON F. GOTTLIEB
. 3. 4. 5.
........................................................... ..........................
1 Introduction 2 . Some Specific Uses of Hyperbaric Oxygenation
......
Limitations Associated with Pressure and Hyperbaric Oxygenation Analytical Techniques .................................................. Conclusions ............................................................ References ................................................................
69 75
94 115 119 121
Determination of Hemoglobin and Its Derivatives
E . J . VAN KAMPENAND W . G . ZIJLSTRA Introduction ........................................................... The Standardized Determination of Hemoglobin ........................ . Absorption Spectra and Extinction Coefficients .......................... General Aspects of Two-Wavelength Methods .......................... Spectrophotometric Determination of the Oxygen Saturation ........... . Spectrophotometric Determination of Carboxyhemoglobin ............... Spectrophotometric Determination of Hemoglobin ...................... Spectrophotometric Determination of Sulfhemoglobin ................... References ...............................................................
1. 2. 3 4. 5. 6 7. 8.
xi
142 146 155 165 167 170 176 182 185
xii
CONTENTS
Blood-Coagulation Factor VIII: Genetics. Physiological Control. and Bioassay
..
G I C. INGRAM
. 2.
Introduction ........................................................... References. Section 1 ................................................. Genetics ............................................................... References] Section 2 .................................................. 3. Physiological Control .................................................. References] Section 3 .................................................. 4. Bioassay ............................................................... References. Section 4 ................................................. 1
189 190 191 205 207 218 222 235
Albumin and ”Total Globulin” Fractions of Blood
DEREE WATSON
. . . .
1 2 3. 4 5. 6
Introduction ........................................................... Separation of Albumin from Globulins .................................. Plasma Albumin and Its Variants ...................................... Hypoalbuminemia in Health and Disease ................................ “Total Globulin” Fraction of Blood .................................... Methods for Determining Albumin and “Total Globulin” ................ References ................................................................
237 238 248 256 261 264 282
AUTHORINDEX .............................................................
305
SUBJECT I N D............................................................ ~
330
LIST OF TABLES Copper Metabolism
by ANDREW SASS-KORTSAK
. . 3.
1 Color Reagents for Copper ............................................ 2 Properties of Useful Radioactive Copper Isotopes .....................
4. 5. 6. 7
.
Copper Content of Human Organs and Tismes ........................ Copper Content of Biological Fluids in Man ............................ Copper Proteins in Mammals ......................................... Changes of Human Serum Copper and Ceruloplasmin Levels with Age Metabolic Abnormalities in Symptomatic and Early Asymptomatic Homozygous Patients and Heterosygotes of Wilson's Disease ........
5 13 16 19 21
34 46
Determination of Hemoglobin and Its Derivatives
by E. J . VAN KAMPEN AND W . G . ZIJLSTRA
. . 3. 4. 5. 6.
1 Values of DEoN and of ~ZC~N. Based on Different Determinations of CHION. for 21 Hemoglobin Preparations (Z1) 2 Average and Standard Deviations of &!aN, Based on Different Deter-
7. 8
. 9. 10.
. .
11 12
13. 14
.
.
16
.
16
........................
minations of CHiON .................................................. Values of Light Absorption Maxima and Minima of H b .............. Isobestic Points (IBP) of H b with HbOz, HbCO, and Hi; e Values of H b at the Isobestic Wavelengths ................................... e Values of Light Absorption Maxima and Minima of HbOz .......... Isobestic Points (IBP) of HbO, with Hb, HbCO, and Hi; e Values of HbO, at the Isobestic Wavelengths .................................. e Values of Light Absorption Maxima and Minima of HbCO ........... Isobestic Points (IBP) of HbCO with Hb, HbO,, and Hi; e Values of HbCO at the Isobestic Wavelengths ................................ e Values of Light Absorption Maxima and Minima of Hi .............. Isobestic Points (IBP) of Hi with Hb, HbOz, and HbCO; e Values of H i at the Isobestic Wavelengths .................................... Light Absorption of HiCN Solutions in the Near Infrared .............. e Values of Light Absorption Maxima and Minima of HiCN and the Value of e " / P ..................................................... Comparison of the Two Methods Described in Section 5.1 ............. Comparison of SO, as Measured by Van Slyke Analysis and by Spectrophotometric Measurement a t h = 560 and 522 nm ................... Effect on SCOwhen Diluting a Blood Sample Containing HbCO and HbOz ............................................................... Effect of Saturation with Oxygen on So0 as Determined with the Methods Described in Sections 6.1.1 and 6.2.1 ................................. xiii e
150 151 157 158 159 159 160 160 162 162 163 164 169 171
175 175
xiv
LIST O F TABLES
17. SOOMeasured by Methods 6.1.1 and 6.2.1 in Mixtures of Solutions Containing 100% HbOo and 100% HbCO, Respectively .................. 18. Comparison of SOODetermined Using Methods 6.1.1 and 6.2.1 .......... 19. SHIof 5 Blood Samples as Determined by the Three Methods Described 20. S H ~of 4 Blood Samples as Determined by the Methods Described in Sections 7.2 and 7.3 .................................................
176 177 181 181
Blood-Coagulation Factor VIII: Genetics, Physiological Control, and Bioassay
by G. I. C. INGRAM 1. Graham’s “M” Kindred ............................................... 2. Neutralization Tests of Anti-Factor-VIII by Hemophilic and von Wille-
brand Plasmas
......................................................
199 203
3. Quantities Required in the Calculation of Potency Ratios from Clotting-
4.
5. 6. 7.
Factor Assays Based on Single Readings on Differing Numbers of 2-Fold Dilutions of Patient and Control Plasmas ..................... Formulas Required in Obtaining Slope Indices and Potency Ratios for Assays in Which All Plasmas Are Tested at the Same Numbers of Dilutions ........................................................... Values of Q for i = 6 and i = 10 ...................................... Results Obtained in a Factor VIII Assay .............................. Log Transform of the Results Shown in Table 6 .......................
230
232 233 234 235
Albumin and “Total Globulin” Fractions of Blood
by DEREK WATSON 1. Amino Acid Groupings in Albumins .................................... 2. Main Components of Normal “Total Globulin” of Human Blood Serum
3. Nitrogen Content of Human Plasma Albumin and “Globulins” ........ 4. Chromogenicity of Some Plasma Proteins toward the Alkaline Copper Reagent of Gornall and the Alkaline Copper Phosphotungsto-phosphomolybdate Procedure of Lowry ...................................... 5. Tryptophan and Tyrosine Contents of Some Purified Human Plasma Proteins ............................................................ 6. Serum Protein Values Compared by Densitometry (R20), Refractometry, and Biuret Reaction (512) ..........................................
249 262 265
271 278 281
COPPER METABOLISM
Andrew Sass- Kortsak Department of Paediatrics, University of Toronto, and The Research Institute, The Hospital for Sick Children, Toronto, Canada
1. Introduction and Historical Review.. . . . . ................. ... 2. Methods of Investigation.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Determination of Copper in Biological Materials. . . .. .. . . . . . . . . . . . . . .
.
............ 3. Copper in Tissues and Biological Fluids. . . . . 3.1. Copper Content.. . . . . . . . . . . . . . . . . . . . 3.2. The State of Copper in Tissues; Copper ................. 4. Intake, Elimination, and Transport of Copper., . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Nutritional Aspects. . . . . . . . . . .... .. .. .
............ 5. Physiological and Pathological Variations of
5.3. Effect of Pregnancy and Hormones.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4. Alterations of Copper Metabolism in Pathological States. . . . . . . . . . . . . References . . . . .. . . . . . . . . . . .. . .. . , . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
..
. ...
.
1 3 3 10 13 13 15 15 19 24 24 25 26 28 32 33 35 35 36 50
1. Introduction and Historical Review
Copper was one of the first metals known to man, first used probably between 8000 and 7000 B.C. Copper utensils were found in Egyptian tombs dating back as far as 5000 B.C. The word copper derives from the Latin cuprum, an altered version of cyprium. Cyprium is an abbreviation of aes cyprium, the original name for copper in Latin, and refers to the island of Cyprus where the ancient copper mines were located (Sll). In spite of the long history of association between copper and man, the fact that copper is present in living organisms did not become known until 150 years ago, when copper was found in plants (B31, M18) and soon after 1 During the preparation of this review the author was supported by a grant from the John A. Hartford Foundation of New York. 1
2
ANDREW SASS-KORTSAR
in animal tissues (S6). Nevertheless for some time thereafter it was considered that copper in living organisms, as well as some other trace elements, was a mere contaminant introduced from the environment either in vivo or during the process of sampling. The first indication that copper is an essential body constituent, a t least in some species, came with the recognition of the copper-containing respiratory pigment hemocyanin in cold-blooded animals (FS), and the discovery of turacin, a copper-containing pigment in the feathers of certain birds (C10). Actual nutritional dependence of animals on copper was not conclusively proven until 1928, when Hart et al. (H4) showed copper to be an essential nutrient for the rat and other animal species (E7). The recognition of naturally occurring copper-deficiency states in a number of domestic animals was a further important step toward the establishment of copper as an essential nutrient. Cattle, sheep, and pigs feeding on plants grown on copper-deficient soil were reported to be affected in North America (N2)) in Europe (529, S30), and in Australia (B14). It had long been suspected that copper in the living organism is concerned with “catalytic” functions (G15).With the advent of modern biochemistry it was possible to prove that this is indeed the case. Copper was recognized to form an essential constituent of the molecule in a number of enzymes. In many of these, copper is a part of or essential for the function of the active site of the enzyme. Some of these enzymes, like cytochrome c oxidase, are widespread in the living tissues of a many animal species and indispensably involved in basic life processes. This establishes copper as an essential element for a wide variety of living organisms more forcefully than the evidence from nutritional studies. While trace amounts of copper are essential for life, an excess of copper is toxic. There are many reports on the toxic effects of copper in a wide variety of animals and in man; these are well summarized in Underwood’s monograph on trace metals (U2). Copper is well known as an inhibitor of many enzymes in extremely low concentration, around M . The mechanism of such inhibition may be through an antagonistic effect on other essential metals, by the binding of the copper to the active site of the en~ymeand thus rendering it inactive, and by a tendency of copper ions to combine with essential cofactors such as glutathione, etc. (R7). It seems evident from all this that in living organisms, including man, elaborate physiological mechanisms must function to ensure that copper is made available in sufficient amount and in the right form whenever and wherever it is needed and, a t the same time, that accumulation of toxic amounts is prevented.
COPPER METABOLISM
3
These mechanisms, all part of the field of copper metabolism, will be the subject of this review. An attempt will be made to summarize presentday knowledge in the field from the literature and from the author’s experience. Preference will be given to the observations made in man. The results of experimental work in animals and of studies in vitro will be used to fill in the gaps. Emphasis will be placed on recent contributions and on the few human pathological conditions related to disturbances of copper metabolism. Because of space limitations and the plethora of recent developments in the field, some important lines of information must be omitted. The reader is referred to a number of relatively recent, excellent monographs and review articles on copper metabolism (Al, A2,C2, S13,S17, U2). Much useful information is also contained in recent reviews and symposia on Wilson’s disease, the most important aberration of copper metabolism in man (B5,S24, WS). 2. Methods of Investigation
Some of the methods of investigation essential for the study of copper metabolism will be surveyed in this chapter. In addition to the analytical methods for the detection and quantitative determination of copper in biological materials, the techniques for determination and detection of one of the biologically important copper proteins (ceruloplasmin) will also be discussed. Finally, the general principles and techniques involved in the use of radiocopper will be surveyed. 2.1. DETERMINATION OF COPPER IN BIOLOGICAL MATERIALS Only trace amounts of copper are present in biological material, in the general range 0.1-10 pg/g (0.1-10 ppm). These trace amounts are in complex mixture with innumerable other metals and elements, some of which are present in much larger quantities. In addition the sample size is often limited, as in blood samples from infants and small animals or needle biopsy specimens from tissues. Methods for copper therefore must be highly sensitive as well as highly specific, since the analyst faces the problem of trace analysis on a micro- or ultramicroscale. Because of the ubiquitous presence of copper in nature and especially in the modern environment, contamination of samples before or during analysis is an ever present danger. For example, ordinary distilled water from copper or tinned copper stills contains quite regularly 10-200 pg/liter of copper. This of course precludes its use for the rinsing of glamware or as a diluent of samples and reagents. Glass-distilled water or distilled water passed through suitable cation-exchange resins must be used, and its quality frequently checked. Glassware must be of highest quality glass, free
4
ANDREW SASS-KORTSAK
from extensive etching, meticulously cleaned and washed in dilute acid followed by extensive rinsing in copper-free distilled water. It is well to remember that filter paper, dialyzing membranes, etc., may contain copper and at the same time may retain copper from the samples. Reagents must be of highest analytical grade, carefully screened for copper contamination. Solvents and concentrated acids may have to be redistilled. As is customary in trace analysis, reagent blanks and standards carried through the whole procedure must be included with practically each batch of determinations. Much useful information concerning the dangers of contamination in trace analysis and its elimination is given by Thiers (Tl). Copper has a marked tendency to form complexes with organic substances, especially proteins, polypeptides, and amino acids. In complexed form, copper may not react with reagents or may not be extracted quantitatively. Special care must therefore be taken to free copper from these bonds quantitatively before its determination. This often cannot be achieved without dry or wet ashing. Numerous methods are available for the determination of trace amounts of copper. The statement has been made, “The methods now available are in the main very satisfactory and leave little to be desired” (53). While this may be true in general, there is still room and need for improvement. Simplification of existing methods would reduce the danger of contamination, and increase in sensitivity would allow the use of smaller samples. 2.1.1. Colorimetric Methods
These methods have the great advantage of not requiring expensive apparatus or highly specialized technical knowledge for operation of the apparatus. The sensitivity range of present-day colorimetric methods is adequate for most purposes; their specificity and accuracy are also considerable. As a rule, preliminary ashing to destroy organic material is necessary. Dry ashing, even at temperatures not higher than 5OO0C,is less satisfactory than wet ashing. This is not so much because of the danger of volatilization of copper, but because of its partial conversion to acid-insoluble forms by reaction with constituents of ash or the vessel wall (S3).Wet ashing with sulfuric and nitric acids or perchloric acid is preferred because there is less danger of loss of material; however, there is danger of contamination unless the acids used are of the highest degree of purity. The use of perchloric acid carries the danger of explosion in inexperienced hands. There are a great number of color reagents for copper, some of which are listed in Table 1. Their relative sensitivity can be compared on the basis of their molar absorbancy indices. The references given in the last column of Table 1 pertain to the sensitivity, specificity, and general charac-
TABLE 1 COLORREAQENTS FOR COPPER
Reagent
Solvent
Wavelength (md
Reference
435
8,000 12,700 14,7OOa
(G14, P5) (E3, M4, P5) (B17)
cc4
435
16,2000
(B17, G5, M13, M17, V2)
Dithizone 2,2Biquinolie (cuprobe)
CCl, Isoamyl alcohol
508 540
24,600
(W
5,900”
(B17, H18, P5)
2,9-Dimethyl-l,lO-phenanthroline (neocuproine) 4,7-Diphenyl-l, 10-phenanthroline 2,9-Dimethyl-4,7diphenyl-l,l0-phenanthroline (bathocuproine)
CC14-EtOH
454
8,000
(J5, P5, 532)
Biscyclohexanoneoxalyldihydrasone (cuprizone)
Diethyldithiomrbamate
440
Molar absorbancy index (1 cm)
Water Isoamyl alcohol CCl,
440
Dibenzyldithiocarbamate
Oxalyldihydrazide
a
+ acetaldehyde
Calculated from data available in reference (B17).
1 d
E2i m
0
12,000 14,200
(P5, 534)
n-Hexanol
420 480
Water
600
16,000
(P5)
Water
542
29,500
(G10, R11, S37)
Water Benzene
540 540
158,000 55,000
(s32, 533, Z1)
(T8) (546)
El-z
6
ANDREW SASS-KORTSAK
teristics of these reagents and describe applications of copper determination in biological materials. Each reagent has its proponents, and the relative merits are difficult to judge from the rather voluminous literature. The most time-honored and certainly dependable method for serum, urine, and tissue copper determination is that described by Eden and Green in 1940 (E3). Wet ashing with a combination of sulfuric, perchloric, and nitric acids is followed by the deionization of iron with citrate and/or pyrophosphate in strongly alkaline medium. Diethyldithiocarbamate is the color reagent and the complex is extracted into amyl alcohol. Provided blanks can be kept low, which can be achieved only by applying extreme care to avoid contamination of the many reagents, copper can be determined reliably in 0.5-1.0 ml of serum. While this method undoubtedly represents a great improvement over previous methods (M4, T5), it is somewhat cumbersome; nevertheless it has proved its worth in many hands and has been the basis of a number of modifications and applications. Gubler et al. (G14) have described another method for the determination of copper in blood by dithiocarbamate. They found ashing unnecessary, and recommend liberation of copper from its organic complexes by incubation with hydrochloric acid preliminary to precipitation of the proteins by trichloroacetic acid. Citrate and pyrophosphate are added, followed by alkalinization with ammonia, as in the previous method. Diethyldithiocarbamate is used as color reagent. The color is read in the aqueous solution, extraction into organic solvent being omitted. The method can be applied to whole blood as well as plasma. This method has not been used as widely as the former. Recovery of protein-bound copper without digestion may not be complete. As can be seen from Table 1, dithiocarbamate produces less color in aqueous solution than in organic solvents. Also, in water the copper-dithiocarbamate complex is in colloidal suspension, which makes colorimetric readings uncertain (P5, S3). The diethyldithiocarbamate copper complex is unstable in acid medium and is light-sensitive. Dibenzyldithiocarbate (available as zinc salt) lacks these disadvantages; moreover, it is more sensitive (Table 1) and more specific (M13, 53). Maytum et al. (M17) and most recently Giorgio et al. (G5) have described methods for urinary copper determination with this reagent. The latter method does not require preliminary ashing. The only disadvantage of these methods is the use of carbon tetrachloride, which is potentially toxic. Dithizone has been used widely in the past for the determination of copper; however, there is little to recommend its use now. It is less sensitive (Table 1) and far less specific than other reagents.
COPPER METABOLISM
7
2,2-Biquinoline and 2,Q-dimethyl-l,lO-phenanthroline are very specific reagents for copper but lack sensitivity (B17); nevertheless the latter has been used successfully for the determination of copper in certain foodstuffs (J5). 4,7-Diphenyl-l,lO-phenanthroline and bathocuproine are more sensitive; the latter is specific for copper to such an extent that it can be used for the simultaneous determination of both copper and iron in serum (21). Rice (R11) and Welshman (W9) have described methods for the determination of serum copper using oxalyldihydrazide (see Table 1) as the color reagent, based on Stark and Dawson’s (537) recommendations for the use of oxalyldihydrazide and the Gubler et al. (G14) method of liberating protein-bound copper without ashing. These methods are simple, claimed to be accurate, and more sensitive than previous methods. Morell e l al. use another method for determining microgram quantities of copper in biological materials.2 The sample is wet ashed by a method similar to that of Eden and Green (E3), and dicyclohexanoneoxalyldihydrazide is used as the color reagent (see Table l), recommended by Peterson and Bollier (P5).Morell’s method takes few short cuts and seems to combine relative simplicity and ease of operation with high sensitivity. An ultramicro modification of this method is used for the determination of copper in needle biopsy specimens of the liver. Finally, Stoner and Dasler’s method utilizes the most sensitive color reagent for copper, l15-diphenylcarbohydrazide (546). Wet ashing is included in this method, followed by performance of the color reaction and extraction of the colored complex into specially treated benzene. This method seems most promising and may prove to be more sensitive than the others. The above methods have all been, or could easily be, adapted for the chemical determination of copper in tissues. Histochemical demonstration of copper is possible with the aid of rubeanic acid (U3). This method, applied to a needle biopsy specimen, has been suggested as a useful diagnostic aid in Wilson’s disease ((78). Unfortunately, the method is rather erratic and does not function in all hands. Improvements in this technique or new, more sensitive methods with high optical resolution for histochemical demonstration of copper are necessary. Of special importance would be the demonstration of copper at the electron microscopic level; unfortunately, such methods are not yet available.
* The latest version of this method has not been published, but can be obtained from the authors on request (A. G. Morell, Albert Einstein College of Medicine, Yeshiva University, New York 61, New York).
8
ANDREW SASS-KORTSAK
2.1.2. Emission Spectrography Emission spectrography has been applied fairly extensively for trace metal analysis in biological material. Early methods were semiquantitative. More exact quantitation has become feasible lately by determining the relative intensity of the photographed spectrum lines by densitometry. The method is particularly suitable for parallel determination of large numbers of elements in the same sample. Butt et al. (B34), Koch et al. (K13), and Tipton et al. (T3, T4) have reported human tissue analysis for copper and a great number of other elements by this method. Long (L9) has reported the use of emission spectroscopy for the detection of metals in serum protein fractions separated by starch block electrophoresis. The technique of emission spectrometry has recently been improved by the use of a direct reading attachment which replaces the more cumbersome and less accurate photographic recording methods. Nusbaum et al. (N8) reported the use of this latter method for tissue analyses, and Valberg et al. (Vl) for red blood cells. The spectroscopic method for the determination of copper is both sensitive and specific; its only disadvantage is the need for complicated and expensive apparatus, which will preclude for some time its widespread use. Another form of emission spectrometry is flame spectrophotometry. Newman and Ryan (N5) have adapted the method for the determination of copper in urine and serum. This method is less sensitive and requires wet ashing as well as the extraction of copper as a dithizone complex into carbon tetrachloride followed by oxidation of dithizone and re-extraction of the copper into water. 2.1.3. Atomic Absorption Spectrometry Willis (W12) has recently summarized the principles and applications of this method. A short note appeared recently regarding the use of atomic absorption spectrometry for serum and urine copper analysis (B15). The sensitivity of this method for copper is rather less than for such other biologically important trace metals as magnesium, zinc, and sodium. The sensitivity can be improved by extracting the copper as dithiocarbamate or pyrollidinedithiocarbamate complex (A7) into methyl isobutyl ketone. While this method is less sensitive than some others, it is nevertheless very specific and the apparatus is only moderately expensive. 2.1.4. X - R a y Spectroscopy Alexander (A6) determined copper, iron, and zinc in dry ashed tissues or plasma by an X-ray fluorescence method. As it stands now, 20-250 mg
COPPER METABOLISM
9
of tissue or plasma ash is necessary as a single sample. Perhaps this amount can be reduced, but until this occurs the method cannot be considered a microtechnique. The high degree of specificity and the need for only dry ashing and plating to prepare a sample for measurement are great advantages of this method. The apparatus is expensive and not yet easily available. 2.1.5. Neutron Activation Analysis
This is presently the most promising method for the determination of trace amounts of copper on an ultramicroscale. The radiochemical properties of copper are such that one theoretically should be able to detect 10-10 g of this element without much difficulty (531). There are two naturally occurring nonradioactive copper isotopes, Cua3and Cun6.Exposure to a high flux of thermal neutrons converts these to radioactive Cua4and Cum,respectively. Most authors use Cue4 for detection; Kaiser and Meinke (Kl) see advantages in the use of CuN with a half-life of 5.1 minutes. Their recovery, however, using rat tissues for analysis, is not more than SOY0,which seriously limits the value of the method. When using C d 4it is best to measure the 0.511-MeV annihilation radiation. The main problem in the case of biological samples is the production of 15-hr half-life NaZ4during activation. The back-scattered radiation from the large 1.37-MeV photopeak of Na24 completely obscures the relatively small photopeak of Cua4.For this reason the copper must be separated from the sodium, which may precede the neutron activation or, more conveniently, be carried out following the activation step. I n the latter case, carrier copper can be added, which means that the separation can be carried out on a macroscale and of course contamination with inactive copper is of no importance whatsoever. Kanabrocki et al. (K2) described a neutron activation technique for the determination of copper in biological fluids in which sodium is separated by dialysis, before activation. Bowen (B20) uses a chemical precipitation technique to separate the copper following activation. Kjellin (K8) and Fritze et al. (F10) use ion-exchange chromatography as the method of separation following neutron activation to determine copper in cerebrospinal fluid and in small samples of animal tissues. The great advantage of the chromatographic method of separation is a 100% yield in terms of copper, which cannot be said of the chemical separation techniques. At present activation analysis is the most sensitive method for copper determination. However, it requires the analyst to have access to a nearby source of thermal neutron radiation, as well as a special laboratory and personnel for radiochemical work.
10
ANDREW SASS-KORTSAK
2.1.6. Electrometric Titration Johnson (53) has reported a relatively simple method for determining copper in serum by electrometric titration. OF CERULOPLASMIN 2.2. DETERMINATION
Colorimetric, enzymatic, and immunochemical methods are available for the determination of ceruloplasmin. 2.2.1. Colorimetric Method The colorimetric method is based on the intense blue color of this protein, which disappears under the influence of high concentrations of ascorbic acid or cyanide. The absorption maximum a t 610 mp and the absorbancy index were first established by Holmberg and Laurel1 (H14). Scheinberg and Morel1 (S16) described a technique by which ceruloplasmin can be determined in ceruloplasmin-rich purified plasma fractions. It is said that this method with slight modifications can also be used for the determination of ceruloplasmin in citrated plasma. Due to the relatively low concentration of ceruloplasmin in plasma, however, this method is impractical since it requires large amounts of serum and is disturbed by the slightest turbidity change. It has nevertheless been used to standardize the enzymatic methods and has great value in this respect. 2.2.2. Enzymatic Method This method of ceruloplasmin determination is based on the weak polyphenol oxidase activity of ceruloplasmin, the only oxidase of this kind in human and animal plasma. The best known substrate (in vitro) is p-phenylenediamine or N,N-dimethyl-p-phenylenediamine.The rate of oxidation can be followed by measuring the oxygen uptake manometrically or, more conveniently, by measuring colorimetrically the rate of appearance of the purple reaction product. The exact nature of this reaction is not known. Rice suggested that Bandrowski's base is formed through the condensation of 3 molecules of pphenylenediamine (R12). Colombo and Richterich (Cll), however, were unable to confirm this. The recent work of Peisach and Levine (P4) suggests that Bandrowski's base is a t best only an intermediary product. The end product of the reaction is unknown. The enzyme reaction is influenced by many factors and therefore rigidly controlled conditions must be observed in order to obtain results which reflect only the variation in enzyme concentration. Several well described methods are available (C11, R4, Sl6). The pH optimum of the reaction is somewhere between 5.0 and 6.0,
COPPER METABOLISM
11
depending on such factors as the choice of buffer, molarity, etc. An interesting initial lag period of the enzyme reaction has been noted by several authors (H16, H20, P4);the reason for this lag is at present unknown. Humoller et al. (H20) have suggested that the ascorbic acid present in the serum will reduce the p-phenylenediamine oxidized by the enzyme until it is used up in this reaction. This is disputed by Colombo and Richterich (C11). The lag period is more likely explained on the basis of the complex nature of the oxidation of p-phenylenediamine which proceeds in several steps, according to the recent work of Peisach and Levine (P4).Whatever the explanation, the lag period must be taken into consideration when measuring ceruloplasmin activity by colorimetry. It may not suffice to determine the optical density of the reaction mixture at the start of the reaction and once later. Because of the lag period, especially if there is a short time-interval between the two observations, the straight line between the two points will not represent the rate of the reaction for obvious reasons. A recording photometer will measure the true rate of the reaction. Certain metals have a profound influence on the activity of the enzyme. Copper in concentrations as low as those resulting from contamination of glassware and reagents will catalyze nonenzymatic oxidation of p-phenylenediamine. For this reason it was recommended that 10-3M EDTA (ethylenediaminetetraacetic acid) be incorporated in the reaction mixture (B28). However, higher concentrations of EDTA inhibit the enzyme, and it has been suggested that EDTA may form a complex with copper in ceruloplasmin and thereby inactivate the enzyme (H20, K3). The addition of EDTA to the reaction mixture therefore may not be desirable. We found that by simply avoiding copper contamination, by the use of precautions necessary in copper determinations, the nonenzymatic oxidation of p-phenylenediamine slows to an imperceptibly low rate, and in this situation there is of course no need to add EDTA. Levine and Peisach called attention to the effect of Fez+ ions on the reaction (L7). Fez+ ions in low concentration stimulate and in higher concentration inhibit the reaction. Their p-phenylenediamine preparation was the source of ferrous ion contamination. Micromolar concentrations of EDTA abolish this effect by chelating the iron. We feel that p-phenylenediamine free of iron contamination should be used. This will obviate the need for EDTA, since the iron present in serum normally does not disturb the reaction (Cll), probably because it is present in the ferric state (Fe3+) and bound to transferrin as well. Finally, the effect of a number of anions on the oxidase activity of ceruloplasmin should be mentioned. Holmberg and Laurel1 (H16) and later Curzon (C18) showed that chloride, acetate, nitrate, bromide, thiocyanate,
12
ANDREW SASS-KORTSAK
sulfate, and fluoride all have inhibitory effects. Serum albumin was claimed to be inhibitory, but was found later to inhibit only the nonenzymatic oxidation of p-phenylenediamine and not to inhibit or enhance the oxidase activity of ceruloplasmin (R4). These observations underline the need for meticulous standardization of enzymatic assay techniques for determining ceruloplasmin. With proper standardization, the enzymatic methods are reliable. However, one must always be alert to the possibility that factors other than the concentration of ceruloplasmin may influence the p-phenylenediamine oxidase activity. The results of enzymatic determinations of ceruloplasmin are often expressed in arbitrary units, and the values judged in the light of a series of results obtained in normal subjects by the same method. Expression of the enzyme activity in milligrams of ceruloplasmin per unit volume of serum is also possible. The relation between oxidase activity and the amount of ceruloplasmin in serum can be determined by measuring in parallel samples of sera both the oxidase activity and the change of optical density a t 610 mp before and after the addition of ascorbic acid or cyanide. On the basis of the known absorbancy index, the ceruloplasmin concentration can be calculated (see Section 2.2.1) and the relation between it and the enzyme activity determined. Alternatively, purified human ceruloplasmin can be used for standardization of the enzymatic method. The ceruloplasmin content of the purified preparation can be determined colorimetrically or, in the case of a highly purified preparation, by nitrogen analysis. Predetermined increments of ceruloplasmin can then be added to aliquots of a selected serum. It is convenient to select a serum with relatively low ceruloplasmin level to start with. Serum of a patient with Wilson’s disease, some of whom have no measurable amount of enzyme activity, would be ideal for the purpose; however, Walshe (W5) has recently found an inhibitor in these sera. Mention should be made of a simple enzymatic screening method whereby individuals with very low ceruloplasmin blood levels can be detected (A4). The method employs a filter paper disc (Whatman No. 3). A drop of 0.4% p-phenylenediamine in acetate buffer of pH 5.2,l.O M , and a drop of plasma are placed on the filter paper in succession. The discs are then wrapped in a polyethylene sheet, placed in a tray, and incubated a t 3740°C. A control sample with a borderline low ceruloplasmin level is included among the unknowns. The intensity of the blue color, which develops as the result of ceruloplasmin-catalyzed oxidation of p-phenylenediamine, is noted and compared to that produced by the standard. Convenient kits for the performance of this screening test can be obtained ~ommercially.~ 8
Dade Reagents Inc., Miami, Florida.
COPPER METABOLISM
13
2.2.3. Immunochemical Method Immunochemical methods are also available for quantitative determination of ceruloplasmin (H11, M11, 515). These require a highly purified preparation of human ceruloplasmin, which serves as antigen and is given parenterally to another species, usually rabbits, to incite the production of specific antibody against the human ceruloplasmin. The specificity of the antibody depends entirely on the purity of the antigen. Further purification of the antibody can be achieved by absorption on sera devoid of ceruloplasmin, e.g., sera of certain patients with Wilson’s disease. The immunochemical method of determination of ceruloplasmin yields results which correlate well with serum copper levels and with the results of the enzymatic method. OF NON-CERTJLOPLASMIN-BOUND COPPERIN SERUM 2.3. DETERMINATION
A small but important fraction of serum copper is not bound to ceruloplasmin. Most of this “non-ceruloplasmin-bound copper” is attached to serum albumin. The amount of this fraction of serum copper varies in physiological and especially in pathological conditions. It can be measured since it will react with diethyldithiocarbamate directly, whereas ceruloplasmin-bound copper will react only after acidification. A method for the determination of “direct reacting” copper in serum has been published by Gubler et al. (G12). Unfortunately, this method is not very accurate, especially in the case of normal subjects where the figures are very low. More exact measurements can be obtained when the values are higher as in certain pathological states. COPPER 2.4. USE OF RADIOACTIVE There are two useful radioactive isotopes of copper. Certain properties of these are shown in Table 2. TABLE 2 PROPERTIES OF USEFULRADIOACTIVE COPPERISOTOPES’ Property Half-life Energy of principal
y
emission
Maximum energy of fl emission
a
cum
CUE’
12.8 hr 0.511 MeV (from p+ decay)
58.5 hr 0.184 MeV 0.093 MeV
,T 0.57 MeV
j3- 0.40 MeV
j3+ 0.66 MeV
0.48 MeV 0.58 MeV
Data were taken from reference (Rl).
14
ANDREIW SASS-ICORTSAK
Copper-64 is readily available commercially, and so far has been more widely used than Cus7. Its half-life is short and it therefore can be used only for relatively short-term experiments. By virtue of positron emission, it yields two rather energetic y-rays through an annihilation reaction which can be detected and measured with relatively high efficiency. Copper-67 (not available commercially) has been produced and used for some experiments (544).Its half-life is longer than that of Cuss and it is also readily measured by its y-emission. The energy peaks of CuB4and Cua7 are conveniently different, which permit their use in double-labeling experiments (544). Both CuB4and Cue?emit @-radiation;thus they can be measured by @-countingtechniques and detected by autoradiography. Unfortunately, their maximum @-energyis rather high and the autoradiographic resolution therefore is rather poor, which impairs their suitability for microautoradiography. Copper-64 of a medium degree of specific activity can be produced by bombarding Cum,one of the stable copper isotopes, with thermal neutrons to yield CuB4by the following reaction. CU"
(n,y) CUM
Very high specific activity, in fact almost carrier-free CuB4,can be produced by fast neutron bombardment of Zna4. ZnM (n,p) CUM
This is followed by chemical separation of the copper from zinc (F9). Lots containing 20-25 mC of activity in less than 1 pg of copper have been produced by this method. Copper-67 can be produced from Ni64 (an enriched preparation of which is necessary) by a beam of accelerated a! particles from a cyclotron. NiM
( c Y , ~ Cue7 )
The resulting CuS7is contaminated to some extent with Cus4. Copper-67 can also be produced in a reactor by fast neutron bombardment of Zns7.An enriched preparation of Zna7is necessary. The reaction is Zne7 (n,p)Cue7
and is analogous to the production of CuB4from ZnB4.Here again chemical separation of the Cue' from Zna7is necessary. The specific activity of the resulting preparation is expected to be high. In vitro incorporation of radioactive copper into ceruloplasmin is possible
COPPER METABOLISM
(S44).Ceruloplasmin tagged with purposes (Kl).
15
has also been used for experimental
3. Copper in Tissues and Biological Fluids
3.1. COPPERCONTENT
The copper content of the normal human adult is widely quoted as 100-150 mg. This figure, as far as we could determine, is taken from a paper published in 1935 by Chou and Adolph (C9). These authors measured the copper content of 9 organs taken a t autopsy from 2 adults, one of whom died of opium poisoning, the other of a ruptured aortic aneurysm. The total copper content of these organs was calculated with figures for water content and weight of the organs from standard tables. The approximate total body copper of 115.9 mg was obtained by adding the total copper content of the 9 organs. The authors state that their figure is only a rough estimate since brain, skin, and other tissues were not included; nevertheless they suggested, “Probably the total copper content (of the human body) is between 100 and 150 mg.” Tipton and Cook (T3) recently have given the results of a quantitative emission spectrographic analysis of 24 trace elements in 29 human tissues. Tissues were obtained from 150 adults from various parts of the United States, who were free of disease and died suddenly, usually as a result of accident. Copper was one of the elements studied, and, since these data constitute the most complete material so far obtained by the most carefully standardized methods of collection and analysis, the results are shown in Table 3. In the original paper, the copper content is given as pg/g ash. For easy reference and since tissue copper levels in the literature are usually given on a dry or wet weight basis, we have calculated these with the aid of ash as per cent of the dry and wet weights given by Tipton and Cook in their paper. The median and the 80% range are shown in the table for each organ, expressed as pg/g ash, d r y weight, and wet weight. Using the organ mass values established for the “standard man” (R9), the median and 80% range of total copper content were calculated in each organ and are shown in the last column of Table 3. The data in Table 3 speak for themselves. It is interesting to note that liver, brain, heart muscle, and kidney have the highest concentration of copper, in decreasing order. There is a higher concentration of copper in the upper parts of the gut than in the lower parts. Skeletal muscle has a relatively low copper level; nevertheless, because of its large mass, muscle as a compartment contains more copper than any other tissue. One-third of the total body copper is contained in muscle. The copper content of
CL
Q,
TABLE 3 COPPERCONTENT OF HUMAN ORGANS AND TISSUES (1) pg
Orgllns and t i i e a
Brain Liver Kidney Heart Diaphragm Muscle
Skin
Omentum Spleen PsnereaS Lung
Lanmx
Trashes EsophsguS stomach Gut
Duodenum JejWlu,
n m cemun
Siioid Rectum Adrenal Thyroid Testis Prostate hsry
Uterus Uriuaw bladder Aorta
(2)
Cu/g ash-
Cu/g dry weight
~rg
pg C u / g
(31 wet weightb
M W
(4)
(5)
mg Cu in whole organd
Median
80%range
Median
80%rwe
Median
sO%range
(9)
Median
So% range
360 510 260 350 140 74 100 190 89 140 120 33 56 120 230
240-500
21.9 18.8 11.9 14.0 4.76 3.11 1.70 0.38 4.45 4.90 5.75 3.07 3.02 4.56 7.36
14 3-30.5 11.8-48.0 8.75-15.6 10 .0-18.4 3.74-6.80 2.10-5.05 1.36-2.55 0.22-0.56 3.35-6.00 3.22-7.35 4.41-8.65 1.30-13.9 1.35-5.19 3.19-7.60 4.80-10.2
5.40 6.63 2.86 3.85 1.34 0.89 0.70 0.38 1.25 1.68 1.32 0.99 0.90 1 .08 1.84
3.60-7.50 4.16-16.9 2.W 3 . 7 4 2.75-5.06 1.06-1.92 0.60-1.44 0.56-1.05
1.500 1.700 0.300 0.300
8.10 11.28 0.86 1.16
5.40-11.25 7.07-28.70 0.83-1.11 0.83-1.52
30.000 2.000
0.22-0.56
lo.m
0.94-1.68 1.12-2.52 1.02-1.98 0.42-4.50 0.4Lk1.54 0.76-1.80 1.20-2.56
0.150 0.070 1.000
26.70 1.40 3 . w 0.19 0.12 1.32
18.OM3.20 1.12-2.10 2.2&5,6@ 0.144.25 0.08-0.18 1.02-1.98
0.365
0.39
0.28-0.66
260 230 240 220 210 170 200 94
160420 160-330 17M00 150-280 15fk340 110-270 160-250
10.5 8.97 8.16 4.40 5.25 5.27 2.20 3.48 4.55 4.50
6.25-16.4 6.25-12.9 5.78-13.6 3.00-5.60 3.75-7.00 3.41-8.38 1.76-2.75 1.59-8.88 3.30-6.27 2.56-6.30 3.74-7.91 3.14-5.59 1.75-4.15 1.90-5.32
2.08 2.07 1.75 1.32 1.47 1.19 0.92 1.03 0.88 1.10 1.16 0.96 0.88 1.26
12 8 3 . 3 6 1.44-2.95 1.2w.04 0.90-1.68 1.05-2.38 0.77-1.89 0.74-1.15 0.47-2.64 0.64-1.21 0.63-1.54 0.82-1.75 0.72-1.27 0.61-1.44 0.70-1.96
2.OOO
3.34f
1 ,13-2.55f
0.02
0.008
0.02 0.035 0.022 0.009
0.015-0.023 0.004-0.053 0.026-0.048 0.0134.031 0.0074.014
0.150
0.132
0.0914.216
so
100 120 98 110 90
320-1300 190-340
250-460 110-zoo 50-120 80-150 110-280 67-120 92-210 92180 14-150 25-96 84-Ux)
150-320
43-240 58-110 57-140 Sl80 73-130 76-180 50-140
5.28 4.21 2.53 3.42
-
-
0.020 0.020
0.040 0.020
-
-
-
-
-
-
-
-
-
Bloodg Bone (vertebra)h Subcutaneous t h e , bone marrow, lymphoid tissue
}
Estimated total body content:
-
-
-
5.0
2.0-5.0
1.45
0.581.45
0.98 0.75
0.72-1.24 0.30.75
5.400 7.000
5.290 5.250
3.898.70 2.1-5.25
-
-
-
-
-
-
7.800
6.57
4.96-8.92
-
76.0
50.0-120.0
From Tipton and Cook (T3), except where indicated otherwise. Calculated from data given by Tipton and Cook (T3).except where indicated otherwise. 0 From Report of I.C.R.P. Committee I I a s ) . d Calculated from data in columna 3 and 4. Since omentum is mainly depot fat the copper content of depot fat was calculated here based on the total maas of fat in the body (shown in column 4 ) and the copper concentration in the omentum (column 3). I Calculated on the basis of the arithmetic mean copper content of the various pa& of the gastrointestinal tract. D Concentration of copper in whole blood of man waa taken from Wintrobe et al. (W15). Calculations were based on mean f 2 standard deviations. h Bone copper contents, p e r s o d communication from Dr. Isobel Tipton (1965). 4 There are no data in the literature on the copper content of these t h e a . The assumption was made therefore that the copper content of subcutaneous tissue irr the same aa that of skin, and that the copper content of bone marrow and lymphoid tissue is the eame as that of blood. a
b
0 0 'd
z
z
18
ANDREW SASS-KORTSAK
human hair and nails (not shown in Table 3) is considerably higher than that of other organs, compared on a wet weight basis (M14, R13), no doubt due to the low water content of these tissues. Copper in hair and nail does not reflect the accumulation or depletion of copper in the body, thus the analysis of these easily accessible tissues is of little value for diagnostic purposes (M14, R13). On the basis of the excellent data of Tipton and Cook (Table 3) with a few assumptions, which are detailed in the table, the total body copper content can be calculated. The mean of 75 mg with a range of 50-120 mg is lower than the previously quoted figures. The average whole-body concentration of copper in adult vertebrates including man is given as 1.5-2.5 pg/g fat-free tissue (W10, Wll). If 75 mg is taken as the total body content, calculated from Tipton’s data (Table 3), and 60 kg as the fat-free average weight of a human subject (R9), the mean concentration of copper in the whole body of man is 1.25 pg/g fat-free tissue with a range of 0.83-2.0 pg/g. These figures are lower than those given by Widdowson et al. (W10, Wll). With a few exceptions, the relative distribution of copper in the tissues of animals and of man, is not very different (U2). Sheep and cattle (ruminants in general) have a higher concentration of copper in the liver than other animals (Al). The distribution of copper in organs and tissues has been studied by histochemical methods. In the liver most of the copper is in the parenchymal cells (H19, U3). Because of the low sensitivity of histochemical methods, this could be demonstrated only when the amount of copper in the liver was in excess of normal. There is wide variation in the copper content of various parts of the human brain. Gray matter in general has a much higher copper content than white matter. The locus coeruleus, a small bluecolored area in the brain, contains a surprisingly high amount of copper, 107404 pg/g dry weight (El). Copper can be demonstrated in nerve cells by histochemical methods only when their copper content is abnormally high (H19). By chemical methods, which are more sensitive, the presence of copper in white matter has been proven conclusively (El). The intracellular distribution of copper has been studied. A distinct pattern of distribution was found among the four subcellular fractions that can be separated by differential centrifugation. In rat liver, 64.3% of the total copper is found in the soluble fraction, 8.2% and 5.0% in the mitochondria1 and microsomal fractions, respectively, and 20.3% in the fraction containing nuclei and cell residue. These results obtained by Thiers and Vallee (T2) were confirmed by Hennann and Kun (H10). I n the livers obtained at autopsy of two adult men, Porter has found a similar subcellular distribution of copper (P11, P15).
19
COPPER METABOLISM
The concentrations of copper in biological fluids of man are shown in Table 4. The blood levels of copper are around 100 pg/lOO ml. Plasma and erythrocytes contain similar concentrations of copper. The copper content of saliva is much lower than that of the blood. Bile contains highly variable amounts of copper, which is not surprising since it is the main excretory route for copper. The copper content of cerebrospinal fluid is controversial. According to the recent careful measurements of Kjellin (K8) (shown in Table 4), the copper concentration in the cerebrospinal fluid of normal adults is very low. Higher levels were reported by others (see literature quoted by Kjellin, KS). TABLE 4
COPPERCONTENT OF BIOLOGICAL FLUIDS IN MAN Copper Fluid
(pg/lOO ml f S.D.)
Reference
Blood Whole Plasma Erythrocytes Saliva Bile Cerebrospinal fluid
Urine
Very little copper is excreted in urine. Here the concentration values have little meaning; the 24-hour excretion is shown in Table 4. The figures in this table were taken from the careful study of Butler and Newman (B33). Bearn and Kunkel (B7) and Porter (P9) found similarly low urinary excretions. According to our own experience, the 24-hour urinary copper excretion does not exceed 40 pg in normal individuals. Others accept figures up to 60 or even 100 pg per 24 hours as normal (see references quoted in reference B33). 3.2. THESTATEOF COPPERIN TISSUES;COPPERPROTEINS The physicochemical properties of the element copper are such that free ionic copper probably does not exist in appreciable amounts in the living organism. An exception may be the stomach contents where a relatively high degree of acidity may allow the solution of copper ions. In the rest of the body, copper is in a complexed, more or less tightly bound form with proteins, peptides, amino acids, and Probably other organic substances.
20
ANDREW SASS-KORTSAK
Many, perhaps all, proteins have some degree of copper-complexing property. Of the serum proteins, albumin and transferrin have known copper-binding properties. When, however, copper is added to serum in excess of a 1 :1 molar ratio in respect to albumin, then, in addition to albumin and transferrin, several other globulins will also bind some copper. While many proteins in tissues may thus easily form copper complexes, there are a few proteins of which copper is an integral part. Copper in these proteins is part of the molecular structure and not in dissociation equilibrium with ionic copper in the solution; there is a characteristic ratio between moles of protein and atoms of associated copper. These and a few other criteria, discussed in detail by Vallee (V3), make these proteins a special class of metal-protein complexes and establish them as metalloproteins. An ever increasing number of copper metalloproteins is being recognized. Those regularly present in mammals are listed in Table 5 with some of their characteristics. Other important naturally occurring copper proteins, such as hemocyanin, laccase, and ascorbic acid oxidase, are not listed since they do not occur in mammals. The metalloprotein nature of some of the proteins listed in Table 5 has not been established fully as yet. The search for further copper proteins, copper-protein complexes, and other forms in which copper may be stored or transported in the body must continue. Many of these copper proteins are enzymes, and the copper is a part of their active group (Nos. 1, 10-14 in Table 4), while others have no known enzyme activity (Nos. 2-9). As far as is known, none of these proteins functions as a respiratory carrier, as hemocyanin does in mollusks. It has been suggested, but not proven, that the “human liver copper protein” of Morell et al. (M32) and the “hepatic mitochondrocuprein” of Porter et al. (P13, P15) may function as copper storage proteins, similar perhaps to ferritin in the case of iron. It may be questioned whether all these proteins are separate entities. There are many similarities between hemocuprein and erythrocuprein, and also between hepatocuprein, horse liver copper protein, and human hepatocuprein. Porter suggested that the erythrocuprein of Markowitz et al. (P14), cerebrocuprein I, and human hepatocuprein may be identical proteins, although he recognizes some differences between them. The copper in some of these copper proteins accounts for a sizable portion or all of the copper present in the organ compartment where they are found. Copper locked in ceruloplasmin accounts for 90-95a/, of the copper in plasma (H14). Copper bound to “human liver copper protein,” according to Morell, accounts for as much as 80% of the total copper in the subcellular soluble fraction of liver (M32), and another 7% is accounted for
TABLE 5 COPPERPROTEINS IN MAMMALS Molecular weight NO.
__
Protein
9
Ceruloplasmin Hemocuprein Erythrocuprein Hepatocuprein Horse liver Cu protein Human liver hepatocuprein Human liver Cu protein Hepatic mitochondrocuprein Cerebrocuprein I
10 11 12
Cytochrome c ox:dase Tyrosinaae Uricase
13
Monoamine oxidase
14
8-Mercaptopyruvate transsulfuram
1 2
3 4 5 6
7 8
Source
Color
(9)
cu Cu content (gram-atom Enzyme (%) per mole) activity
Human plasma 160,000 Blue Bovine erythrocytes 36,000 Blue Human erythrocytes 28,00&33,200 Colorless Bovine liver (not determined) Colorless Equine liver 30,000-40,000 Bluish green
0.34 0.34 0.324.3fi 0.34 0.344.41
Human liver
(not determined) Green
0.3
Human liver
-10,Ooo
Human and bovine (not determined) neonatal liver Human and bovine 30,00&40,000 brain 93,000 Bovine heart Melanoma, s k i 33 ,000 Tissues of mammals 100,000 except primates 255 .000 Vertebrate tissues
Rat liver
10,000
8 2 2
+
2
Not found Notfound Not found Notfound
Unknown Unknown Unknown Unknown Unknown
-
Not found
Unknown
-
Colorless
0.1-2.7
(not known)
2.4
-
Bluish green
0.29
2
-
0.07 0.02 0.056
1
Colorless
-
0.07
4
+ + + +
-
0.82
1
+
-
Physiological role
Reference (H14. K3, P3) (M10) (K7. M11. 525) (M10) (Mz7) (P14)
(variable) Not found Cu storage?
(M32. 525)
Not found Cu storage?
8 1 3 , P15)
Notfound
Unknown
(ma, P12)
Electron transport Phenol oxidase Urate oxidation
(A9.55, W1. Y2) (B18, F6, P8) (M5)
1 1
Oxidativedeamina- (Y1) tion of cateohol aminea and others Transsulfuration (FI) of &mereaptopyruvate
Y
22
ANDREW SASS-KORTSAK
by Porter’s hepatocuprein (P14). In human red blood cells, 60% of the copper is in the form of erythrocuprein; the rest is bound to another protein which has not been characterized as yet (S26). The last five copper proteins in Table 5 account for vitally important but very small portions of the copper in tissues. hluch work has been done with recently available electron spin resonance and nuclear magnetic resonance techniques to: (a) confirm the presence of copper in a protein, (b) elucidate its valence state in the protein, and (c) show a possible change in the valence state of copper on addition of substrate. These aspects are beyond the scope of this review and are well summarized, together with other recent developments in the general field of metalloproteins, in a recent comprehensive review by Malmstrom and Keilands (M7). I t was pointed out earlier, but bears repetition here, that the mere existence of these copper proteins and the indispensablefunction of some of them are perhaps the best proof that copper is an essential nutrient for man. Ceruloplasmin, the copper protein in plasma, deserves special attention. Human plasma contains approximately 32 mg/100 ml of this protein (see Table 6) (C2, C12, R4). According to the recent careful measurements of Kasper and Deutsch (K3), the molecular weight of ceruloplasmin is 160,000, somewhat higher than the 151,000 obtained by Pedersen (P3). Ceruloplasmin is an az-globulin and contains 8 atoms of copper per molecule (H14). In addition to being a metalloprotein, it is a glycoprotein containing 7% carbohydrate : hexose, hexosamine, and neuraminic acid (U).Ceruloplasmin has been prepared in a pure form from human and pig sera (H14). Several newer methods are now available for its isolation and purification (B29, C20, D4, L4, 54, 57, S39). Ceruloplasmin has an intense and characteristic blue color, with a strong absorption maximum at 610 mp (H14). The extinction coefficient in a 1-cm cell is E$$ = 0.68 (K3, S16), which is very similar to the extinction coefficient that can be calculated from the data in Holmberg and Laurell’s paper describing the first isolation of ceruloplasmin (H14). Ceruloplasmin is an enzyme exhibiting oxidase activity against several substrates with a pH optimum between 5.4 and 5.9. The best known substrate is p-phenylenediamine or its dimethyl derivative. The oxidase activity is much weaker against other substrates such as hydroquinone, catechol, pyrogallol, DOPA, adrenaline, noradrenaline, serotonin (U).The physiological substrate of ceruloplasmin, if any, has not yet been found. The blue color and the enzyme activity of ceruloplasmin disappear when copper is removed from the molecule or when the protein is treated with reducing substances. The change in color upon the addition of reducing
COPPER METABOLISM
23
substances is attributed to a change in the valence of copper (L4). The 8 copper atoms in ceruloplasmin do not have identical properties. Scheinberg and Morell (516) found that 4 of the 8 atoms of copper are exchangeable with ionic copper in the presence of ascorbic acid. Curzon showed that, by digestion of ceruloplasmin with chymotrypsin, 4 atoms of copper can be liberated per mole (C17). By electron spin resonance studies (B16, B30, K4) and by magnetic susceptibility measurements (MS), it was established that 4 atoms of copper in ceruloplasmin are in the cupric state, and the other 4 are in the cuprous state. The blue color and the enzyme activity are dependent on the presence of the cupric atoms (B16). The color is not dependent on the presence of molecular oxygen in the ceruloplasmin molecule (M33). The nature of the copper-protein bond in ceruloplasmin is not known. Copper-binding agents, such as diethyldithiocarbamate at pH 5.7-8.8 (L4, S16) or EDTA at pH 5.5 (L7), do not remove copper from ceruloplasmin. At lower pH, however, in addition to the above two, a number of copper-chelating agents will remove copper (L4). Lowering the pH below 4.75 alone leads to loss of the characteristic blue color (G12, H14). This effect recently has been shown to be reversible; thus, under certain conditions upon raising the pH there is very definite, although not complete, return of color and enzyme activity, and the typical hyperfine structure in the electron spin resonance spectrum reappears (C19). A reversible change in the nature of the copper-protein bond seems to have been produced under these conditions. Morell and Scheinberg (M30), employing conditions under which exchange between ceruloplasmin and ionic copper was shown to occur (S16), successfully prepared an apoprotein of ceruloplasmin, which contained only 4 atoms of copper per molecule, had no color, and was devoid of enzyme activity. Recombination of the apoprotein with copper occurred in the presence of ascorbic acid (h130). Further studies of this apoprotein and similar apoproteins produced by exposure of ceruloplasmin to Tris-hydrochloric acid buffer at pH 7.4 were recently reported by Kasper and Deutsch (K3). They found these apoproteins to have the same molecular weight as the native ceruloplasmin, but to show an increased asymmetry of the molecule, which suggests that some or all of the 4 atoms of copper that were removed play an important role in t,he maintenance of the secondary or tertiary structure of ceruloplasmin. One may speculate that the mere treatment of ceruloplasmin with acid leads to partial unfolding of the molecule by affecting the copper-protein bond insofar as it is involved in the maintenance of tertiary or secondary structure and that copper-binding agents and reducing agents then have easier access to copper attached to the unfolded molecule. ScheinEerg et al. (SlS)
24
ANDREW SASS-KORTSAK
suggested that 16 free carboxyl groups, probably in aspartyl or glutamyl residues, are the sites of copper binding in ceruloplasmin. Kasper and Deutsch (K3) have recent evidence against this contention. Their results suggest that histidyl and either tyrosyl or lysyl residues or both are involved in the binding of copper in ceruloplasmin. It has been suggested that the ceruloplasmin molecule is heterogeneous, as are haptoglobins, transferrins, and certain other proteins (B28, M1, M31, R15, R16,57). A genetic basis for this has been proposed (M31), but could not be proven. It may well be that the observed heterogeneity of ceruloplasmin is the result of partial denaturation, polymerization, or dissociation of the protein into several nonidentical subunits rather than the existence of several molecular species (K3, P16, P17, P18, R15). 4. Intake, Elimination, and Transport of Copper
4.1. NUTRITIONAL ASPECTS The average North American diet contains 3-5 mg copper per day. Because of the ubiquitous presence of copper in food constituents and even in drinking water, it is difficult to devise a balanced diet composed of natural foods that contains less than 1 mg copper per day. The daily minimal requirement of copper for the adult man is stated to be 2 mg per day (C5). Infants require 0.05 mg/kg body weight per day (Nl). These figures are only approximate and most probably far too high, since copper deficiency has not been produced with much lower intakes of copper (W3). Copper balance studies recently performed by Kleinbaum (K9) in fullterm newborn infants showed minor over-all loss or only a minimal degree of copper retention. Premature infants, without exception, were found to have a slightly negative copper balance. These findings suggest that in this age group there may not be a dependence on dietary supply of copper. The increased copper depots present in newborn and premature infants are apparently being used preferentially for growth requirements in this age group. Full-term infants, after the first month of life up to 1 year, without exception showed positive balances: a retention of 0.1-0.4 mg copper per day (K9). This indicates a need for dietary copper. Not knowing the efficiency of absorption from the bowel, one cannot predict nutritional requirements from these studies. Naturally occurring copper deficiency has not been proven to occur in man, most probably because of minimal requirements and plentiful supply in the diet. The copper deficiency, suspected on the basis of low serum copper levels in infants with various types of protein-losing enteropathy (548, 23), is of doubtful pathognomonic significance.
25
COPPER METABOLISM
Copper deficiency occurs naturally and can be produced experimentally in a number of animal species. The chapter on copper in Underwood’s “Trace Elements in Human and Animal Nutrition” is an excellent source of information pertaining to this subject (U2). OF COPPERFROM 4.2. ABSORPTION
THE
INTESTINE
Thompsett (T6) remarked in 1940 that the study of copper absorption seems to have been neglected. He reports a single experiment on copper absorption in mice and concludes that acid facilitates, high calcium intake inhibits, and low calcium intake facilitates absorption. Gitlin et al. (G6) found in mice that the higher the oral dose of copper the more was absorbed. Their results suggest the existence of two mechanisms for absorption: a mechanism following first-order kinetics, and a second mechanism suggestive of an enzymatic process. Crampton et al. (C15), using everted sacs of hamster intestine, found maximal transport of Cu64from the mucosal to the serosal side in the lower middle part of the small intestine. There was little increase in the rate of transport above the initial concentration of 1 pg/ml of copper in the medium. Under anaerobic conditions transport was much reduced. Findings in intact dogs suggest that maximal copper absorption occurs in the upper parts of the small intestine (Sl). Radioactive copper given by mouth to human subjects appears very rapibly in the blood (B9, C5, J l ) , which indicates rapid absorption, probably from the upper parts of the small intestine and stomach. There is evidence to suggest that the form in which copper is present in food greatly influences its availability for absorption. Mills has shown that copper in herbage is in the form of water-soluble, negatively charged organic complexes, and that copper administered in this form is more effective in increasing copper stores in the copper-deficient rat than inorganic copper sulfate (M21-M23). It was suggested that copper may in fact be transported through the intestinal mucosa in the form of a negatively charged organic complex as well as in ionic form (M24). While absorption of the copper contained in foods of plant origin is seemingly efficient, it has been reported that rats fed on raw meat developed copper deficiency. This suggests that the copper present in raw meat is not avaliable for absorption. Experience in this direction is well summarized and the pertinent literature quoted in a recent article by Moore et al. (M29). Guggenheim (G16), on the other hand, presents recent evidence to the effect that the “meat anemia” in mice is due to the high zinc content of meat in the presence of low but, by itself, adequate amounts of copper, and that a concomitant lack of calcium further aggravates the situation. One must wait to see whether this applies to the raw meat copper deficiency of
26
ANDREW SASS-KORTSAK
rats, and it must be explained why copper deficiency does not develop on a diet containing cooked meat, as Moore et al. (M29) have shown. A great deal remains to be learned about the effect of the form of copper in the diet on its availability for absorption. The absorption of copper from the bowel can be influenced by a number of inorganic substances. Anions, which have a tendency to form highly insoluble salts with copper under the conditions prevailing in bowel contents, tend to reduce the absorption of copper. Sulfide is the best known of these. Cupric sulfide is poorly absorbed by rats and pigs (B21, 523). Addition of sulfides to the diet markedly reduces copper absorption in animals (D5) and man (C5). Of the cations, molybdenum is known to influence copper retention in animals. It is suspected that molybdenum influences the membrane transport of copper in such a way that absorption is decreased and excretion increased (D6). For details of this problem the reader is referred to Underwood’s monograph (U2) and a recent review by Mills (M25). Problems of this nature have not yet been encountered in man. Oral doses of potassium sulfide and certain ion-exchange resins are used to reduce the absorption of copper in certain pathological conditions in man (C5). Although certain factors influencing the absorption of copper from the intestine are recognized, we have no concept of the mechanism of absorption nor do we know the regulatory mechanisms which influence the rate of absorption. The suggestion of Scheinberg and Morel1 (516) that ceruloplasmin may be involved in the regulation of copper absorption could not be substantiated. 4.3. EXCRETION OF COPPER
The main route of excretion of copper from the body is via the feces. Urine contains extremely small amounts of copper (4-30 pg/day in man) (B33). Sweat contains only negligible quantities (M26). There is very little copper in saliva, and it is recirculated. Insignificant amounts are lost by menstruation (L6). The amount of copper in feces depends to a great extent on the dietary intake (K9). Copper output via feces is not a reliable measure of excretion of copper from the body since some of the dietary copper may pass through the gastrointestinal tract without being absorbed and thus may constitute a significant portion of fecal copper. It has been shown, however, by several investigators that copper given by the intravenous route is excreted very efficiently and almost exclusively via feces. Gitlin et al. (G6) found that, short of giving lethal doses of copper intravenously to mice, the higher the dose the larger the amounts excreted in the feces. In man, up to 33% of an
COPPER METABOLISM
27
intravenous dose of radioactive copper can be recovered from the stools, and only a few tenths of a per cent of the dose appear in the urine (B9). Most of the copper excreted from the body via the faces reaches the intestine by the bile. In dogs, 7% of an intravenous dose of C U M was excreted in t,he feces; 81% of this came through the bile, 13% passed through the intestinal wall directly, and only 0.5% in the urine (M6). In rats, Neumann et al. (N4) recovered 7% of a 3-pg intravenous dose of copper in the bile during 10 hours following the injection. Owen (05) reported more recently that, following the intravenous administration of 5-100-pg doses of labeled copper to rats, 20-50% of the injected dose was recovered in the bile during 24 hours following injection; the higher the dose, the greater the percentage recovered in the bile. Figure 1 is a scintillation scan of the upper abdominal region of a normal human subject who received 24 hours earlier a l-mg intravenous dose of copper labeled
FIG.1. Scintillation scan of the upper abdomen of a control human subject, taken 24 hours after the intravenous administration of 1 mg radioactive labeled copper. The triangular shadow represents the liver. The dots along the curved line below the liver represent radioactive material in the second part of the duodenum.
28
ANDREW SASS-KORTSAK
with CUM. The dark areas represent concentrations of radioactive material. The liver is outlined very clearly. In addition, a semicircular area of concentration of radioactive material can be seen, which most probably represents activity in the second and third parts of the duodenum. During the next 24 hours, this individual excreted 16% of the radioactive dose via the feces. I t seems likely that the radioactive copper passed into the duodenum via the bile, probably following a sudden contraction of the gall bladder. It is hard to explain otherwise the presence of a high concentration of activity in such a localized segment of bowel. Ceruloplasmin can be detected in the bile (J2), but the amounts found cannot account for more than a very small fraction of the copper in bile. Approximately 50% of the copper in bile is bound to various other proteins, and the rest is probably complexed to smaller organic molecules (H5). Hawke and Holmes (H6) found convincing evidence of an enterohepatic circulation of copper. To decide whether copper, once excreted via the bile, could be reabsorbed from the intestine, they cannulated the bile duct of a rat and connected it with a tube to the duodenum of a second rat. Following intravenous injection of radioactive copper into the first rat, activity could be detected in the blood, bile, and organs of the second rat. Obviously, the Cua4excreted via the bile of the first rat was absorbed from the bowel of the recipient animal. 4.4. TRANSPORT OF COPPERAND DISTRIBUTION FOLLOWING ORAL OR INTRAVENOUS ADMINISTRATION
Copper absorbed from the gastrointestinal tract is transported by the blood to the tissues. Following the ingestion of radioactive copper, there is a rapid increase in plasma radioactivity followed by a rapid fall (B9, B32, C5, E2, J1, M15). Copper administered intravenously as a single dose disappears extremely rapidly from the blood (Fig. 2). Half of a 0.1-l-mg dose of copper disappeared from the plasma compartment in less than 10 minutes. Bearn and Kunkel (B8) found that following the ingestion or intravenous administration of CuM, during the early stages, the Cu" is bound exclusively to serum albumin. I t seems that the portion of serum copper which serves as an immediate transport form of copper and is in equilibrium with tissue copper is not the large compartment (90-95%) bound to ceruloplasmin, but the 5-10% of copper loosely bound to albumin. Ceruloplasmin apparently does not fulfill a function analogous to that of transferrin. It has been shown that exchange between ceruloplasminbound and ionic copper does not occur in the living organism, and that copper is incorporated into ceruloplasmin when the protein is formed and is released only in the course of decay of the protein molecule (544). Con-
COPPER METABOLISM
29
100.0 50.0
0
30.0
M.B. 1 O O p g C u M.G. 80 p g Cu
x
I.W. 1 m g c u
a3 .-G
10.0
I.
a"
.n
0
5.0
R \
o
\
2 0
b
.
,
1
.
1
I
I 0 2 0 30 4 0 5 0 6 0
I
80
Id0
1
I20
T i m e in M i n u t e s
FIG.2. Disappearance of Cue4 from the plasma compartment following intravenous injection of 100-pgto 1-mg doses of radioactive labeled copper i n three human controls. The percentage of the dose remaining in the plasma compartment is plotted against time following injection. sidering this together with the findings of Bearn and Kunkel (B8), outlined in the previous paragraph, it is apparent that the albumin-copper complex is the form in which copper is transported in the blood. Moreover, Neumann and Sass-Kortsak (N3) have shown recently that, in equilibrium with albumin-bound copper, there is a third fraction of serum copper bound t o amino acids. This fraction is rather small compared even to the albumin-bound fraction, but by virtue of its low molecular size it may have a n important function in the transport of copper through membranes. Of the 21 amino acids present in human blood, histidine was the strongest binder of copper in competition with albumin. Some of the
30
ANDREW SASS-HORTSAK
evidence indicated that the true transport form of copper may be a complex formed by 2 different amino acids and 1 atom of copper. However, the isolation and chemical characterization of such complexes have not yet been achieved. Whether amino acid-copper complexes act as carriers facilitating the transport of copper to the cell membrane where acceptor sites with higher affinity for copper take over or whether copper is transported actually through the cell membranes as an amino acid-copper complex remains to be seen. In model experiments with dialyzing membranes, it was shown that, when an acceptor for copper (albumin) was present on both sides of the membrane and CuMwas added to one side, the presence of amino acids greatly facilitated the transfer of copper through the membrane. Much remains to be learned about the actual transport mechanisms of copper from the blood to the tissue cells and through the tissue cells. Returning now to the distribution of copper in the body, a large percentage of the copper which rapidly disappears from the blood is taken up by the liver. Bush el al. (B32) were the first to discover the concentration of activity in the liver following administration of radiocopper. Osborne et al. (01,04) in control human subjects found 6f3-95y0 of an intravenously injected 30-300-pg dose of copper in the liver 2-8 hours following injection. Results of a similar study (58) in a control individual are shown in Fig. 3. The lower part of the graph shows that 86% of the intravenous l-mg dose of copper was concentrated in the liver 4 hours after its administration. Thereafter, activity over the liver declined, a phenomenon also noted earlier by Osborne and Walshe (01). Two routes are available for the discharge of copper from the liver. One is excretion via the bile, for which circumstantial evidence is presented in Fig. 1; this is a scintillation scan obtained in the same normal individual whose results are shown in Fig. 3; it shows the concentration of radioactivity in the duodenum 24 hours after administration of the dose. This individual excreted 16% of the dose via the stool, most of which reached the bowel via the bile. The latter conclusion is probably correct on the basis of evidence presented in Section 4.3. The second route of discharge of copper from the liver is via ceruloplasmin. Ceruloplasmin is synthesized in the liver (05); in the course of synthesis copper is incorporated into the newly formed ceruloplasmin which is then discharged into the blood stream. It was shown by others earlier (B8, B9, B32, 51) and is demonstrated in Fig. 3 that in response to the administration of a single dose of CuMthe initial rapid fall of activity is followed by a marked and sustained secondary rise in total plasma activity. During the first few hours following its administration, the Cus4is exclusively in the albumin-bound form; during the secondary rise of activity, however, most of the CUM is bound to cerulo-
COPPER METABOLISM
31
100.0 60-0
--
40.030.020.0
*
$ 8
-
Total Plasma
100.0-' 60.0-
40.0-
10-o,, , , ,
0 4 8 12
, 18
I
1
2 4 3 0 36 4 2 48
5 4 60 66 72
I
Time after iniection, hours
FIG. 3. The distribution of a single intravenous dose of 1 mg radioactive labeled copper (Cue4)in the various body compartments of a human control. The percentage of the dose in the total plasma compartment, that bound to ceruloplasmin, that taken up by the liver, and that taken up by the rest of the body are plotted against time following injection.
plasmin (B8). The secondary rise of total plasma activity is due to the increasing appearance of ceruloplasmin containing labeled copper, most probably discharged from the liver. During the course of the experiment shown in Fig. 3, the per cent of the dose contained in the liver decreased from 86% to 45%; thus there was a loss of approximately 40% of t'he dose from the liver. Fecal excretion,
32
ANDREW SASS-KORTSAK
probably via the biliary route, accounted for 16.4% of this, and by the end of the study a further 13.2% of the dose was in the circulating plasma, bound to ceruloplasmin. It seems that 30% of the 40% of the dose discharged from the liver is accounted for by the two routes. This is reasonable agreement, considering some margin of error in the results, as well as the fact that fecal reabsorption may have occurred and that some labeled ceruloplasmin most probably equilibrated into the extravascular pool of ceruloplasmin and thus escaped measurement. Total body scintillation scanning following a single intravenous dose of radiocopper has not disclosed significant areas of concentration of activity except in the liver and, to a minor degree, in the kidney (S8). The latter is also evident from the earlier work of Maytum et al. (M17). In spite of a degree of concentration of activity in the kidney, only an insignificant part of this is excreted in the urine. In the course of the experiment shown in Fig. 3, only 0.6% of the dose was excreted in the urine. If an even distribution of the radiocopper in the body is assumed, except for the area of liver and kidney, and after correction of surface counts for activity in the circulating blood, it is evident that some of the injected radiocopper in the study shown in Fig. 3 was t,aken up by the tissues in the rest of the body. This uptake is transient, and the rapid rise of activity is followed by a more gradual decline (lower part of Fig. 2). It is probable that the copper from the tissues during this phase is gradually transferred to the liver. Ceruloplasmin, with the copper locked in the protein, has a much slower turnover than the rest of the circulating copper. The half-life of ceruloplasmin in man was estimated as 4-7 days (S20, S42). Thus the turnover rate of this protein is faster than that of albumin or y-globulin. Ceruloplasmin, as was mentioned previously, is most probably synthesized in the liver; however, the site of its degradation is not known. A most interesting recent suggestion by Broman (B27) ascribes to ceruloplasmin the function of a special copper carrier which transports copper from the liver to the various sites where cytochrome oxidase is synthesized. It is proposed that, “An elaborate prosthetic group comprising four cuprous-cupric pairs framed in a system of ligands may be transferred from the large carrier part of the ceruloplasmin molecule (which is catabolized) to the cytochrome oxidase unit under construction.” This is a most intriguing suggestion that deserves further study. 5. Physiological and Pathological Variations of Copper Metabolism
From the foregoing considerations it is obvious that, in spite of considerable effort on the part of many investigators, the basic processes and mechanisms whereby copper is absorbed, transported, stored, built into body
COPPER METABOLISM
33
constituents, and excreted by the living organism are by no means completely understood. How the various physiological and pathological factors lead to apparent alterations of copper metabolism is even less clearly defined. It is not possible for this reason to place these factors in logical order or to explain fully their effects or mechanism of action. 5.1. EFFECT OF AGE AND SEX ON COPPERMETABOLISM The copper content of fetal and neonatal tissues is much higher than in adults. This is well documented in man (F2, G1, G2, K11, N7) and in many animal species (U2). The most complete data on the copper content of human fetal tissues were reported recently by Fazekas et al. (F2). They measured the copper content of 29 different tissues in a total of 109 human fetuses of differing degree of maturity, including full-term newborn infants. The increased amounts of copper in fetal tissues are more impressive when compared on a dry weight basis, no doubt due to the higher water content of fetal tissues. If Fazekas’ data in full-term newborn infants and the values of Tipton and Cook (T3) in adults on a dry weight basis are compared, it is evident that all newborn tissues contain higher amounts of copper. Liver, muscle, skin, and adrenals are particularly high, 18-20 times the mean adult concentration. The copper concentration of thyroid, testes, and uterus is over 10 times that of adults. The brain copper content, however, is only 2 times the adult’s. Only in the liver is a gradually increasing concentration evident with increasing degree of maturity. In other tissues the copper concentration does not seem to change with age of the fetus. These increased copper depots in the newborn are obviously derived from the mother through the placenta. The copper content of the placenta is rather high (4-5 pg/g wet weight and 3040 pg/g dry weight) (F2, K11). The high tissue copper levels of the newborn decrease soon after birth. In three infants between 3 and 12 months of age Nusbaum et al. (N7) found the copper content of the liver and kidney to be within the normal adult range. On the other hand, Gerlach (Gl) found high liver copper content in some infants up to 1 year of age. In children between 7 and 13 years of age the tissue copper concentrations were not different from those in adults (Gl, S9). The serum copper and ceruloplasmin levels vary quite markedly with age, as shown in Table 6. The mean levels and range of variation are well established in normal adults. The results obtained in three large series of subjects by three different groups of investigators show very close agreement. It has been known for some time that the serum copper content of newborn infants is much lower, about one-third of the normal adult range, and, more significantly, one-sixth of the values obtained in parturient
34
ANDREW SASS-KORTSAK
mothers (H9, H15, 52, S19). The low copper levels in the newborn are paralleled by similarly low serum levels of ceruloplasmin (Table 6). The neonatal levels of the small albumin-bound fraction of serum copper, which can be calculated as the difference between copper bound to eeruloplasmin (0.34%) and total serum copper content, are not significantly different from the maternal levels (S19) or from normal adult levels (C2). The placenta does not seem to permit the transfer of ceruloplasmin from mother to fetus, or only to a very small extent, and perhaps most or all of the fetal ceruloplasmin is synthesized by the fetus. The increased copper stores in CHANGES
OF
TABLE 6 HUMAN SERUM COPPER AND CERULOPL4SMIN LEVELSWITH AGE Copper (pg/lOO ml)
No. Parturient mothers Newborns (cord blood) 2 years 6 years 10 years Adults (21 years) Adults Adults
12 12 133 95
235
I
216 (118-302)' 36 (12-26)" 140 (95-186)' 129 (83-174)b 117 (72-162)' 109 (69-150)'
_ -
114 (89-147)'
Ceruloplasmii~ (mg/100 ml)
No.
12 12 134 136 100 30
Ref.
55.5 ( 3 9 . 6 8 9 .O)a 6 . 5 (1.8-13.1)42.6 (31.P53.9)' 38.1 (26.9-49.2)b 33.5 (22.4-44.6)* 30.8 (20.6-41.0)b 32.3 f 4 . g C 33 (25-43)b
1
Range. 95% limits. 6 Standard deviation. 9
the newborn are apparently transferred from the mother via the albuminbound fraction of serum copper, which seems t o be in equilibrium on both sides of the placenta. The equilibration is probably further aided by the third, small-molecular size amino acid-bound fraction of the serum copper which has recently been shown to exist (N3). While this seems most plausible, the possibility cannot be excluded that ceruloplasmin is transferred from the mother through the placenta and that its rate of breakdown in the fetus exceeds the rate of transfer, therby leading to the marked difference in concentration on the two sides of the placental barrier. Interesting age-dependent changes in the concentration of both serum copper and ceruloplasmin in children were recently found by Cox and Sass-Kortsak (C13). I n a group of 134 healthy children from 2 to 15 years of age with close to equal numbers of males and females, there was no significant difference related to sex, but a highly significant (P < 0.001) negative regression coefficient with increasing age. The figures given in Table 6 for 2, 6, and 10 years of age were calculat#edon the basis of the
COPPER METABOLISM
35
regression lines and their 95% confidence limits. From this it seems that by about 2 years of age the initially rather low levels of copper and ceruloplasmin rise above the normal adult range and then gradually decrease to the normal adult range, which is reached sometime between 13 and 16 years of age. That the method used for measurements is not the determining factor is shown by the fact that both serum copper and ceruloplasmin levels changed in a parallel fashion, and the values obtained by the same methods and the same investigators in adults approximate very closely the values found by two different groups of investigators in similar large series of normal adults. The reasons for and the physiological meaning of the marked changes in tissue copper content, serum copper, and ceruloplasmin levels with age are not known. Cartwright and Wintrobe (C2) reported significantly higher levels of both serum copper and ceruloplasmin in normal adult females compared to males. Cox et al. (C13) in a somewhat smaller series could not confirm this. In the latter series there was no significant difference between the serum copper and ceruloplasmin content of males and females either in children or in adults. 5.2. EFFECT OF DIETON COPPER METABOLISM The copper content of the tissues and blood of animals is markedly dependent on the dietary intake (U2). This has not yet been reported in man. Premature infants could not be made copper deficient by being put on a diet containing very little copper (W13). This, however, may be explained by the presence of large copper depots in this age group (see Section 5.1). The presence of increased amounts of copper in drinking water derived from copper pipes has been shown not to influence the copper content of the liver in man (M2). OF PREGNANCY AND HORMONES 5.3. EFFECT
It is well documented that in normal pregnant women there is a gradual increase of both serum copper and ceruloplasmin levels during pregnancy, culminating in levels approximately twice normal at term (see Table 6) (H9, K15, M12, 52, S19). The copper content of the liver is also increased in pregnancy (Gl). The high levels of serum copper and ceruloplasmin decrease rapidly after delivery to reach the normal adult range within 2 months (H9). The mechanism of the elevation of serum and tissue copper and serum ceruloplasmin in the pregnant woman is not known. Administration of estrogens leads to elevation of serum copper and ceruloplasmin in both man and animals (E6, G4,54,M20, R21, T9). Since there is an increased production of estrogens during pregnancy, it may be
36
ANDBEW SASS-KORTSAK
assumed that this is the main reason for the increase in serum copper and ceruloplasmin. It has been suggested recently that the placenta is essential for the high levels of serum copper in pregnancy, and that hormones (estrogens or others) produced by the placenta are responsible for the alterations of copper metabolism in pregnancy (H7). Administration of testosterone (C14, 54)’ thyroid hormones (M19), and ACTH (E5) has been shown to result in elevation of serum copper levels. The latter effect could have been due to contamination of the ACTH preparation with melanotropin ( M ) .Antithyroid drugs produce a fall in serum copper levels (F7). 5.4. ALTERATIONS OF COPPER METABOLISM IN PATHOLOGICAL STATES More or less constant and in certain instances characteristic alterations of copper metabolism have been reported in a variety of human pathological states. Little is known about the mechanisms whereby the homeostasis of copper is disturbed in these conditions, although logical reasons can be given in a number of instances. These observations are restricted in most cases to a description of the alterations in the blood levels of copper and ceruloplasmin in the patients. Efforts to find an explanation for these changes and attempts to sort out the reasons for their development in the course of the disease are made infrequently. The reason is perhaps the tacit recognition that in most instances the disturbance in copper metabolism does not play a central role in either the causation or mechanism of the disease. There is one notable exception. Hepatolenticular degeneration or Wilson’s disease seems to be based on a genetically determined primary disturbance of copper metabolism. This condition is therefore in a class by itself. Since there is no better classification, we shall discuss in the succeeding three sections the pathological disturbances of copper metabolism under three headings : conditions characterized by hypocupremia, by hypercupremia, and Wilson’s disease. 5.4.1, Diseases Characterized by Hypocupremia Hypocupremia and low blood levels of ceruloplasmin are one of the most characteristic biochemical alterations in Wilson’sdisease. This condition is discussed in a separate section below. Hypocupremia and hypoceruloplasminemia of marked degree are found regularly in patients with kwashiorkor, a form of protein malnutrition with normal caloric intake and complicated by multiple vitamin deficiencies (E4, G9, L2, R8). Macdonald and Warren (M3) found that the copper content of the liver (on a fat-free dry weight basis) is sivignificantly reduced
COPPEB METABOLISM
37
in these children, as compared to values in controls of the same age who died of other diseases. It is not known whether these changes are due to low dietary copper intake, decreased absorption from the gastrointestinal tract, or failure of the synthesis of ceruloplasmin. The finding of low copper stores in the liver seems to favor the former two factors. It is interesting to note that, in rats on a low-protein but normocaloric diet supplemented by essential minerals and vitamins, Allen and Schneiden (A8) found normal copper concentration in the liver. This suggests that copper deficiency in kwashiorkor is not due to the protein deficiency alone, but t o the lack of some other essential dietary constituent, perhaps vitamins or minerals. Low copper and ceruloplasmin levels have been reported in marasmus of infants (protein and caloric malnutrition) from Chile (M28) and India ((39).
Hypocupremia and low serum ceruloplasmin levels were reported in celiac disease (A13, Ll), and in tropical (B36) and nontropical sprue (Cl, S40). Oral administration of copper to the patients with tropical sprue resulted in a sustained rise of serum copper in only one of four patients (B36). Sternlieb and Janowitz (S40) have recently shown that in nontropical sprue the absorption of copper from the gut is impaired. A syndrome consisting of hypocupremia, hypoferremia, hypoproteinemia, edema, and hypochromic microcytic anemia has been described in infants (K10, L1, S48, U1, 23). The hypoferremia, anemia, and hypoproteinemia responded to treatment with iron; however, the hypocupremia required administration of copper as well. Nevertheless, dietary copper deficiency does not seem to be a factor in this condition (W13). These patients are most probably sufferingfrom some form of protein-losing enteropathy, and are losing ceruloplasmin and transferrin along with other serum proteins through the gastrointestinal tract in excess of the ability of the synthetic mechanisms to replace these losses. Hypocupremia is found in protein-losing enteropathy in adults. Low serum copper and ceruloplasmin levels are found regularly in the nephrotic syndrome with hypoproteinemia (C3, M12). These patients were found to excrete 46-75 mg ceruloplasmin per 24 hours in the urine, and this and perhaps the concomitant loss of albumin-bound copper most probably explain the development of hypoceruloplasminemia and hypocupremia in this condition. Slight elevation of serum copper with significant reduction of serum ceruloplasmin levels was found in one large series of patients with multiple sclerosis (P7). What connection this has with the mechanism of the disease is not known. In the majority of patients with liver disease, other than Wilson’s disease,
38
ANDREW SASS-KORTSAK
the serum copper and ceruloplasmin levels are elevated. However, in one young patient with chronic hepatitis, paradoxical low blood levels of both copper and ceruloplasminwere found by Walshe and Briggs (W6); Wilson’s disease was excluded in this patient on solid grounds. Very low ceruloplasmin and copper levels were also found by these authors in four cases of fatal hepatitis. Low serum copper levels were reported in vitiligo (B13). However, the normal series of these authors exhibits a very wide range of variation of serum copper levels with 7 of 25 normal subjects having less than 50 pg/lOO ml of copper. This is most unusual and points to either a methodical error or a peculiar type of population. The mean serum copper level of the group of 76 patients with vitiligo is lower than the mean of the normals subjects; however, statistical proof of the significance of this difference is not given. 5.4.2. Diseases Characterized by Hypercupremia Hypercupremia is more commonly encountered in disease states than hypocupremia. It is well known that, in patients with acute and chronic infections due to a wide variety of viral, bacterial, or parasitic agents, the serum copper levels are elevated, probably due entirely to an increase in ceruloplasmin (B23, B24, H8, H13, L3, M12, R10). Patients with disseminated lupus erythematosus, rheumatoid arthritis, rheumatic fever, and glomerulonephritis have increased blood levels of copper and ceruloplasmin (H8, H15). Other conditions with hypercupremia and hyperceruloplasminemia are myocardial infarction (A3, V2), various types of anemia (C6, H8, L3, P1, P2), lymphogranulomatosis (L3, P1, P2, R5), leukemias (L3, R5), malignant neoplasias of various forms (K5, K14, Pl), and the postoperative state following major surgery (24). The mechanism by which serum ceruloplasmin and copper levelk increase in these conditions is not known. According to one theory, ceruloplasmin is considered as one of the “acute-phase reactants” together with serum glycoproteins, seromucoid, mucopolysaccharides, and C-reactive protein. It is argued that increased blood levels of these, and of ceruloplasmin, are found whenever there is inflammation or other type of tissue destruction in the body. However, the argument in favor of this contention is weakened by the lack of positive correlation between serum ceruloplasmin and C-reactive protein levels in some of these conditions (R10). Elevated serum copper and ceruloplasmin levels were recorded in various forms of acute and chronic liver disease (B10, G13, G17, P6, R2, R6, R14, S9). In a group of patients with portal cirrhosis the serum non-ceruloplasmin-bound copper was also elevated (G13). The elevation of serum copper and ceruloplasmin is most marked in biliary cirrhosis and, generally
COPPER METABOLISM
39
speaking, in liver disease characterized by intra- or extrahepatic biliary obstruction. The blockage of the biliary excretory route of copper is most likely an important factor in the production of hypercupremia in these pa tien ts . Finally, elevated serum copper and ceruloplasmin blood levels were reported in schizophrenia (A5, A l l , B25), and it was postulated that this may be an important pathognomonic feature of this condition. Others, however, were unable to confirm this (F11, H17, S21). I t seems that the higher copper and ceruloplasmin levels reported earlier in schizophrenic individuals can be explained, partly on the basis of a high incidence of acute and chronic infection in institutionalized mentally ill individuals and partly by the presence of low serum ascorbic acid levels which may influence the particular type of oxidase activity determination used in one of these studies (A5). The copper content of tissues has been studied in some of the pathological conditions characterized by elevated serum copper and ceruloplasmin levels (B35, G1, G2, H21, 11, K l l , N6, W16). Moderate but definite increase in the copper content of the liver and of the spleen was reported in chronic infections and malignant diseases, especially with metastases. Marked elevation of the copper content of the spleen was reported in patients with malignant disease complicated by infections (W16). Elevated copper content was also found in the spleen and to a lesser degree in the liver of patients who died of glomerulonephritis (W16). In patients with cirrhosis of the liver, the copper content of the liver and of other organs is quite variable. High liver copper content was reported in some cases, especially in some children with cirrhosis (B35). Unfortunately, Wilson’s disease was not excluded in most of these instances. However, there can be no doubt that in some forms of cirrhosis, especially when biliary obstruction or intrahepatic cholestasis is an important part of the disease process, the copper content of the liver is elevated. For example, in a 6-year-old boy who died with a chronic biliary type of cirrhosis which developed following “neonatal giant cell hepatitis,” we found 240 fig copper per gram (wet weight) of liver, a 20-fold increase over the upper limit of normal. Very high copper content of the liver was reported recently in adults with primary biliiary cirrhosis (H21). Increased liver copper content in the presence of chronic biliary retention is not surprising, since the biliary tract is the main route of copper excretion.
5.4.3. Wilson’sDisease
This condition was first described in 1912 by Kinnier Wilson (W14), an eminent English neurologist. It is characterized clinically by a peculiar
40
ANDREW SASS-KORTSAK
type of nervous system involvement, cirrhosis of the liver, and KayserFleischer rings of the cornea. The disease seems to be based on an inborn error of metabolism, the exact nature of which is not clearly defined, but which leads primarily to a profound disturbance of the metabolism of copper. 5.4.3.1. The Pattern of Inheritance. The familial occurrence of the disease was recorded in Wilson’s first description (W14), and has been confirmed by many others. Bearn’s studies, first on 16 and later on 31 families with Wilson’s disease, proved conclusively the autosomal recessive pattern of inheritance (Bl-B3, B5). Accordingly both parents are as a rule heterozygous in respect to the gene of Wilson’s disease; their children have a 1:4 chance to be homozygous normal, a 1:4 chance t o be homozygous abnormal, and a 1:2 chance to be heterozygotes. There is a high incidence of consanguineous marriages among the parents (B3). The disease is somewhat more prevalent in males. The frequency of the abnormal gene in the American population was calculated to be approximately 1:2000 and the incidence of the disease 1:4,000,000 (B5). This may well be too conservative an estimate. It was suggested that the frequency of the abnormal gene may be higher in the population of southern Italy, among East European Jews (B2), and among Japanese (A12). 5.4.3.2. Clinical Features. The age of onset of the symptoms is highly variable. Late adolescence and the early twenties are the usual times of presentation, but the youngest patient was reported to have first developed symptoms a t 4 years of age (C7) and the oldest during the fourth (536) or even fifth decade of life (S14). The initial neurological symptoms consist usually of clumsiness in the use of extremities followed by the development of tremor, rigidity, dysarthria, and dysphagia. Torsion spasms and abnormal postures may develop. Pyramidal signs may be observed rarely and epileptic seizures may occur. The electroencephalogram is normal until late stages of the disease (B26, 547). In the course of development of the organic neurological symptoms, psychic disturbances in the form of euphoria and other types of inappropriate behavior may appear. By careful testing, latent psychological impairment can be found before the appearance of neurological symptoms and signs (K12). Psychotic behavior occasionally may be the form of presentation of the disease, in which case the erroneous diagnoses of schizophrenia or hysteria often are made. The early hepatic symptoms and signs are those of acute or chronic hepatitis, which then develops into a postnecrotic type of cirrhosis. There is nothing to distinguish the symptoms referable to involvement of the liver in the course of Wilson’s disease from those of postnecrotic cirrhosis
COPPER METABOLISM
41
due to a number of other etiological factors. Liver function tests are generally abnormal, with mild to severe jaundice of the regurgitation type, moderately elevated alkaline phosphatase, abnormal flocculation tests, low blood levels of albumin and elevated yglobulin, low esterified cholesterol, and prothrombin deficiency and other clotting defects. Portal hypertension and bleeding from esophageal varices may develop relatively early in the course of the liver involvement in Wilson’s disease. In addition to the brain and the liver, the kidney is also regularly involved in the disease process, although this is clinically not obvious (B12, H12, W17). A typical clinical feature of the disease is the presence of Kayser-Fleischer rings of the cornea. This is the only important clinical feature of the disease which Wilson missed in his original description. The rings were described, before Wilson’s recognition of hepatolenticular degeneration, by Kayser (K6) and independently by Fleischer (F4) in patients with unrecognized forms of neurological disease and cirrhosis. Fleischer’s case was properly diagnosed later as Wilson’s disease (F5), and the rings were established as typical features of the disease. The rings consist of orange-yellow, sometimes greenish, pigment granules in Descemet’s membrane close to the edge of the cornea; they are sometimes incomplete. During the early stages, the rings may be seen only with the aid of a slit lamp. Their presence is an absolute diagnostic criterion of Wilson’s disease. They may nevertheless be absent during the early stages of the disease, and therefore their absence, especially in young patients, does not exclude the diagnosis. The pigment in Descemet’s membrane contains copper; this has been proven histochemically (B22), spectrographically (G3), and recently by use of the electron probe (T7). A detailed histological, histochemical, and electron microscopic study of the rings was published by Uzman and Jakus (U6). There are other less constant and characteristic clinical features of the disease. Azure lunulae were found in this condition (B11). Severe hemolytic crises were reported as early manifestations in some patients (G11). Some degree of increased destruction of red blood cells may be present throughout the disease, often leading to the formation of pigment stones in the gall bladder. Nonspecific changes in bones were reported by several investigators. These consist of generalized demineralization, spontaneous fractures, pseudofractures, osteomalacia, and even frank rickets (F3, M34, R18). Osteochondritis, periarticular fragmentation of bone, and osteoarthritis may also complicate the disease (F3, M34). While many of the patients develop the full-blown clinical picture of the disease with involvement of the nervous system, liver, and kidney and with Kayser-Fleischer rings, a substantial number may have neurological
42
ANDREW SASS-KORTSAK
manifestations only and no demonstrable liver involvement. Others may present liver disease and die of cirrhosis or its consequences without developing the slightest sign or symptom of neurological involvement. In general, the younger the age of presentation, the more likely is the hepatic form of the disease. Patients with later onset, on the other hand, are more likely to present the neurological form of the disease. An acute and sometimes fulminating form of Wilson’s disease may lead to death in a few weeks to a few months after onset of the first symptoms. This is more commonly seen in children (C8, L5, S28). 5.4.3.3. Pathological Changes. Pathological changes in the organs involved in the disease are numerous, but not specific for the disease. Detailed descriptions of the classical pathological features can be found in a number of earlier publications (D2, H1, W14). In addition to the classical picture of lenticular degeneration, the cerebral cortex and the cerebellum may also be involved, sometimes very severely (S22). The pathological changes in the liver are usually those of postnecrotic cirrhosis and sometimes Laennac’s cirrhosis. Hepatic structural alterations (A10) and finer cytological changes in early stages of the disease were studied recently (G8, 512). Glycogen degeneration of nuclei, deposition of lipofuchsin pigment, and fine cytoplasmic fat droplets were found in early biopsy specimens (S12), but these changes can by no means be considered as specific for the disease. Electron microscopic changes consist mainly of an increase in the number and density of lysosomes, which are more numerous in the pericanalicular zone. Larger cytoplasmic pigment granules are also seen (S12). Schaffner et al. (512) reported striking reduction of acid phosphatase activity by histochemical methods, but Goldfisher (G8) was unable to confirm this. Histological changes in the kidney-consisting of focal areas of degeneration, sloughing, and necrosis of tubular epithelial cells, and in other areas of tubular dilation and separation of the tubular cells from the basement membranewere recently described (W17). 5.4.3.4. Disturbance of Copper Metabolism. A profound and characteristic disturbance of the metabolism of copper is the cardinal biochemical feature of Wilson’s disease. The earliest evidence of this came from Germany, where in 1913 Rumpel (R20) found increased copper and silver content of the liver in patients with so-called pseudosclerosis. Siemerling and Oloff (527) as far back as 1922 suggested that the “pseudosclerosis” of WestphalStrumpell is actually due to accumulation of copper salts in the nervous system, liver, and eye. These early observations were not followed up for over 20 years, until Glazebrook (G7)in 1945 reported high copper content in the liver and brain of a patient with Wilson’s disease, and increased
COPPER METABOLISM
48
excretion of copper in the urine was found 3 years later by Mandelbrote et al. (M9). It is now well established that patients with Wilson’s disease have grossly abnormal amounts of copper in the liver, brain, kidney, and cornea (B19, C16, US). The copper content of other tissues may increase, but much less than in these four organs that are the site of the pathological changes in this disease. Increased urinary copper excretion is also characteristic of the disease; as evidence for this a large number of references to pertinent literature can be found in Cumings’ monograph (C16). Scheinberg and Gitlin (515) and independently Bearn and Kunkel (B7) reported in 1952 that in patients with Wilson’s disease the blood levels of ceruloplasmin are very significantly lowered. This was established by the use of both immunochemical (515) and enzymatic (B7) techniques and has been confirmed by many investigators. With increasing numbers of patients, however, it became obvious that, although the majority of patients had extremely low, sometimes undetectable, amounts of ceruloplasmin in the serum, a significant number of firmly diagnosed patients had half or three-quarters of the normal level (M12). Moreover, a small but significant number of patients with Wilson’s disease were reported more recently to have perfectly normal blood levels of ceruloplasmin (E8, R19, S9). Rosenoer (R17) reported the interesting observation that the initially normal levels of ceruloplasmin in his patient (R19) decreased markedly to very low levels during treatment and efficient decoppering with penicillamine. It has been suggested that certain environmental factors, particularly the increased estrogenic activity which may develop in the course of liver disease, could be responsible for the finding of normal ceruloplasmin levels in some of the patients with Wilson’s disease. However, in two of these patients, normal serum ceruloplasmin levels were present a t a relatively early stage of the disease when no other physical sign of estrogenic overactivity could be detected (S9, SlO). I n another patient, clinical evidence of liver impairment was completely absent (R17, R19). The suggestion (S13) concerning a qualitative abnormality of the ceruloplasmin in a patient with normal blood levels (S9) could not be substantiated by more careful retesting, including members of the patient’s family (S10). Walshe and Carpenter (W7) reported recently that ceruloplasmin in sera of patients with Wilson’s disease responds as an enzyme to certain inhibitors in a manner different from ceruloplasmin in sera of normal individuals. While this finding suggests that the ceruloplasmin in patients with Wilson’s disease is different from normal ceruloplasmin, it is also possible that another factor present in sera of patients with Wilson’s disease and absent
44
ANDREW SASS-KORTSAK
from normal sera is the reason for the altered response to the inhibitors. Thus no conclusive evidence has been presented so far to prove the contention that the ceruloplasmin present in the patient with Wilson’s disease is qualitatively abnormal, although this possibility cannot be excluded. The total serum copper levels are also usually reduced in patients with Wilson’s disease (Bl, C4). Considering the low ceruloplasmin levels, this is not surprising. While the amounts of copper bound to ceruloplasmin are reduced, the non-ceruloplasmin-bound, direct-reacting fraction of serum copper is elevated in these patients ((25). Those with normal levels of ceruloplasmin have a high serum copper, since they have an elevated blood level of non-ceruloplasmin-bound copper together with normal amounts of ceruloplasmin-bound copper (S9). In order to explain the presence of vastly increased copper stores in the tissues of patients with Wilson’s disease, it was suggested that the absorption of copper from the gastrointestinal tract is increased. The finding of markedly positive copper balances in seven patients by two groups of investigators (C4, 22)was used as an argument in favor of this contention. However, the validity of this argument is questionable, since decreased excretion via the bile can equally well explain the experimental data. Whether the patient with Wilson’s disease absorbs copper in excess of the normal amounts is an unanswered question. Studies with radioactive copper have helped to delineate more lucidly the disturbance of copper metabolism in the patient with Wilson’s disease. Numerous investigators have performed such studies (B8, B9, B32, E2, J l , M15, M16, M17, 01, 04,S8, S9, S10, 544).This type of investigation consisted mainly of the oral or intravenous administration of a single dose of radioactive copper, followed by a study of the fate of such a dose in the patients compared to controls. After oral administration, the blood levels of CuMrise more rapidly and subsequently decrease more slowly in the patients than in the controls. The peak blood level is usually higher in the patient. This was also used as an argument in favor of increased copper absorption, but again the results can be equally well explained by decreased rate of removal of CUM from the blood. Following intravenous administration, the disappearance from the blood is slower in patients than in controls, which bears out the point above. The secondary rise of activity in the blood, due to the release of ceruloplasmin-bound Cu6*probably from the liver, is regularly seen in the controls, but never in the patients. The patient with Wilson’s disease seems to be unable to incorporate copper into ceruloplasmin, and, interestingly, this is the case even in those rare patients with normal blood levels of ceruloplasmin (S9).
COPPER METABOLISM
45
The distribution of an intravenous dose of Cu" in the various tissue compartments was studied by several investigators (B32, M17, 01-04, 58). The uptake of Cu" in the liver, the main organ where it is concentrated, is slower than in controls (S8), which parallels the decreased rate of initial disappearance of Cu" from the blood. I n addition, the patient's liver takes up a smaller fraction of the dose than the normal liver (01,02,58). Finally, activity in the liver of controls reaches a peak 4-10 hours after administration of the dose and is followed by a gradual and significant decrease of activity over the liver (Fig. 2), while in the patient with Wilson's disease a gradual slow increase of activity over the liver continues, and throughout the 3-day observation period a peak is never attained and apparently CUM is not being discharged from the liver (01,02, 58). By a scintillation scanning technique (S8) and use of a scintillation probe directed toward the lower pole of the left kidney (M17), a minimal degree of concentration of Cu" can be detected in the kidney in the controls. In patients with Wilson's disease a much higher degree of concentration occurs in the kidney (M17, 58).It was also found in the patients that the rest of the body tissues take up a much higher fraction of the dose than in the controls (58). The pattern and extent of the excretion of Cu" are markedly altered in patients with Wilson's disease. Following both oral and intravenous administration, the urinary excretion of CuMis higher in patients than in controls, whereas fecal excretion is much reduced in the patients (B9, B10, B32, E2). 5.4.3.5. Other Biochemical Abnormalities. Other biochemical abnormalities in Wilson's disease consist of renal aminoaciduria, glycosuria, and sometimes intermittent proteinuria (B6, B12, C12, 538, U4). Increased urinary excretion of urate and phosphate coupled with low uric acid blood levels and hypophosphatemia has also been reported (B12). These abnormalities are considered to be manifestations of renal tubular damage. The clearance of glumse, urate, phosphate, and a-amino nitrogen are increased, and the secretory capacity for p-aminohippurate is decreased (B12). In addition to these manifestations of tubular dysfunction, it was shown that the renal plasma flow and glomerular filtration rate are decreased (B12, H12). The degree of functional renal involvement roughly parallels the severity and duration of the overt disease (B12). 5.4.3.6. Biochemical Findings in Patients during the Early Asymptomatic Phase and in Heterozygotes. A question with both practical and theoretical implications is whether all these biochemical changes are present during the early stages of the disease when the patient is completely asymptomatic and all clinical signs and symptoms of the disease are absent. The practical implication of this question is the obvious importance of early diagnosis in
46
ANDREW SASS-KORTSAK
order to permit the institution of treatment before permanent damage has developed. Theoretically it is important to answer this question, since the earlier a manifestation of the disease is present, the more closely it must be related to the basic abnormality. The findings pertinent to this problem are summarized in Table 7. Findings in heterozygotes are included in a separate column for comparison. TABLE 7 METABOLIC ABNORMALITIES I N SYMPTOMATIC A N D EARLY ASYMPTOMATIC P.4TIENTS AND HETEROZYQOTES O F WILSON’S DISEASE HOMOZYQOUS Homozygotes Abnormality
S e r u m copper Serum ceruloplasmin Non-ceruloplasmin-bound copper Urinary copper Copper in tissues Handling of radiocopper Rate of clearance from blood Incorporation into ceruloplasmin Uptake in liver Initial rate Extent Discharge from liver Excretion Urinary Fecal Other abnormalities Aminoaciduria Phosphaturia low serum P Uricoauria low serum urate
+
+
Symptomatic Asymptomatic Heterozygotes Usually low LOW“ High High Elevated
Usually low Lowa Normal Normal Elevated
Rarely low Rarely low Normal Normal
Slow None
Normal None
Normal Slow
Slow Low Impaired
Normal Normal Impaired
Normal Normal Impaired
High Low
Normal Low
Normal
Present Present Present
Absent Absent Absent
Absent* Absent Absent
?
?
a Occasionally homozygous patients with Wilson’s disease may have normal blood levels of ceruloplasmin (E8, R19, S9). * Amiioaciduria was found in members of 2 families with Wilson’s disease only (S35, U5).
Low serum ceruloplasmin levels, if present, are a constant feature of the disease. Very low levels have been found in completely asymptomatic patients as early as infancy (H3). The author has investigated two completely asymptomatic patients with Wilson’s disease (3 and 21 years of age) and found very low serum ceruloplasmin and reduced serum copper levels. Asymptomatic patients similarly to the symptomatic ones show no evidence of incorporation of Cue4into the ceruloplasmin. The copper content of the
COPPER METABOLISM
47
liver is elevated early, during the asymptomatic phase of the disease. This was found by Harris and Scheinberg (H3) as well as by the author. On the other hand, urinary copper excretion is surprisingly normal in the asymptomatic patient with a normal non-ceruloplasmin-bound copper level in the serum, and the urinary excretion of labeled copper following intravenous administration is not increased. Since in these individuals the initial rate of disappearance of an intravenously injected single dose of Cu" is normal, together with a normal rate and extent of uptake of Cu" in the liver, it seems logical to assume that cupruria and elevated nonceruloplasmin-bound copper levels in the serum are secondary biochemical alterations in Wilson's disease which only appear at a relatively late stage, probably consequent to a high degree of saturation of possible binding sites for copper in the tissues, particularly the liver. An early feature of the disease is the apparent inability to discharge copper from the liver; this is also present to a moderate extent in heterozygotes, but there are not sufficient numbers of observations to consider this proven. On the other hand, manifestations of renal tubular damage are not present in the asymptomatic patient, pointing to the secondary nature of these changes. The finding of abnormalities in heterozygotes is also of both practical and theoretical importance. Low serum copper and ceruloplasmin levels are a rather inconstant finding in heterozygotes and the decrease is not marked (SlO,S35, S43). A far more constant feature is the reduced rate of incorporation of Cu" into ceruloplasmin noted by both Scheinberg's and the author's group (SlO,S43, S45). Osborne and Walshe (01) suggest that the maximum percentage of an intravenous dose of radiocopper taken up by the liver may be lower in heterozygotes than in normal individuals; this has not yet been confirmed. The author found in a few heterozygotes that following intravenous injestion of Cus4the secondary discharge of Cus4from the liver was slower than in the normals. 5.4.3.7. Treatment. Effective methods of treatment are available for the patient with Wilson's disease. The most important are decoppering agents : dimercaprol (BAL) (B4, D3), ethylenediaminetetraacetic acid (EDTA) (B6, C5, M15), penicillamine (S41, W2-W4), and lately dithiocarbamate (549). Adjuncts to the above are the administration of a diet low in copper and potassium sulfide (C5) or certain resins (B6) to decrease copper absorption from the gastrointestinal tract. Of the decoppering agents, penicillamine is both the most effective and the easiest to administer and is therefore the drug of choice. As a rule, patients should be treated continuously with an adult dose of 1 g/day. An important aspect of the treatment is that sometimes the response in the form of clinical improvement
48
ANDREW SASS-KORTSAK
may not be apparent for several months, and therefore reports of failure of this therapeutic approach based on trials shorter than 4-6 months (as a minimum) should be regarded with skepticism. 5.4.3.8. The Nature of the Basic Metabolic Defect. Finally, a few words should be said about the mechanism of Wilson’s disease. There are several theories; however, none has been proven conclusively and the nature of the genetically determined basic defect is still not known. Uaman (11, U5,U7) suggested that the primary disturbance in Wilson’s disease is one of protein metabolism, and that an abnormal protein or polypeptide with a very high avidity for copper is formed in these patients, the disturbances of copper metabolism being secondary in nature. Aminoaciduria and peptiduria, consequencesof the altered protein metabolism, are suggested as the primary biochemical manifestations of the disease. This theory could not gain general recognition for several reasons. The main objection is the failure of the proposer of this theory or others to isolate and characterize the hypothetical abnormal protein and/or polypeptides. Secondly, aminoaciduria and peptiduria were not found to be particularly early manifestations of the disease, nor could aminoaciduria or peptiduria be found with regularity in relatives of patients, except in a family investigated by Uaman (U5)and a family investigated by Soothhill et al. (535). Finally, the theory does not adequately explain many of the characteristic abnormalities of copper metabolism in this disease, in particular the presence of low serum ceruloplasmin levels and deposition of abnormal amounts of copper in the tissues a t a very early age, long before aminoaciduria or peptiduria or other signs of a disturbed protein metabolism are evident. A second general theory which contends that Wilson’s disease is due primarily to a disturbance of copper metabolism is gaining wider recognition. The functional and morphological changes in the tissues and organs involved are regarded as consequences of the toxic action of the excess copper deposited in these organs. It fits in very well that the largest amounts of copper are found in those organs clinically and morphologically involved in the disease process (liver, brain, kidney). The fact that the removal of copper from the patient by the use of decoppering agents is followed by clinical and laboratory improvement is a good argument in favor of this theory. It has also been shown that copper in the amounts present in the tissues will interfere with essential intracellular metabolic processes (H2,
V5). By what mechanism the abnormal amounts of copper are deposited in the body is not known. Scheinberg and Gitlin (S15),finding very low serum ceruloplasmin levels in these patients, suggested that an inability of the patient to synthesize this protein is basically a t fault. Absorption of copper
COPPER METABOLISM
49
from the gastrointestinal tract was suggested to be controlled in some way by ceruloplasmin (Sl6, S17) and in the presence of low blood levels of ceruloplasmin, control over absorption to be lost, and copper to be absorbed in excess of the needs, leading to accumulation of abnormal amounts of copper in the tissues. The ceruloplasmin theory of Wilson’s disease is most attractive but-in this reviewer’s opinion-untenable for several reasons. Patients with normal blood levels of this protein nevertheless died of the disease. It has so far, not been possible to prove that the ceruloplasmin in these patients was in any way abnormal qualitatively. Second, in a number of pathological states other than Wilson’s disease, ceruloplasmin levels may be as low as in some patients with Wilson’s disease without development of clinical or biochemical changes similar to those in patients with Wilson’s disease. It has also been found that patients with Wilson’s disease, during treatment, will show further marked decrease in serum ceruloplasmin levels and, in spite of this, show marked clinical improvement. Finally, the control of copper absorption by ceruloplasmin was predicated on the assumption that an equilibrium exists in vivo between ceruloplasmin-bound and free copper. This, however, could not be proved (S44). A disturbance in the synthesis of ceruloplasmin is no doubt present in most patients with Wilson’s disease; however, it seems to be a n effect rather than the cause of the disease. Evidence is accumulating to suggest that, in addition to the defect in the synthesis of ceruloplasmin, another mechanism whereby copper is discharged from the liver is also disturbed, namely, excretion of copper via the bile. There is only indirect evidence for this, consisting mainly of two observations : 1. Several investigators found normal amounts of copper in the bile of patients with Wilson’s disease (C5, D3). Gitlin el al. (G6) suggested that in the presence of the large excess of copper regularly found in the liver of these patients much higher copper content of the bile should be expected, since this type of response can be obtained in animals overloaded with copper intraperitoneally. Gitlin et al. rightly conclude that the finding of merely normal amounts of copper in the bile of patients who are evidently overloaded with copper must mean that they are not able to excrete copper via the bile under conditions of overloading. 2. Following intravenous administration of Cuffl,patients with Wilson’s disease excrete much less of the Cue4 in the stools than normal individuals (see Section 4.3). Since a high percentage of fecal CUM is excreted into the gastrointestinal tract via the bile, this also suggests that the biliary excretion of copper is impaired in Wilson’s disease. There is no reason why the defect in the synthesis of ceruloplasmin should
50
ANDREW SASS-KORTSAK
influence the excretion of copper via the bile, and vice versa. For this reason it seems logical to assume that the basic defect in Wilson’s disease probably concerns a mechanism by which copper is made available, or presented in a suitably activated form, both for excretion via the bile and for incorporation into ceruloplasmin. Conceivably there are mechanisms of intracellular transport of copper in the liver cell, and a disturbance of these at one point may constitute a block which results in accumulation of copper in the cell and prevents excretion of copper from the cell both via the bile and via synthesis into ceruloplasmin and discharge back into the circulation. The problem, of course, is that next to nothing is known about intracellular transport mechanisms of copper. F-rther experimental work along these lines is needed. 5.4.3.9. The Question of Genetic Homogeneity. The section on Wilson’s disease in this review was introduced by considering the genetic aspects of this condition. Following detailed discussion of clinical, pathological, and biochemical features of the disease, it seems pertinent to return to the problems of inheritance. The question has been rightly asked, “Is the great variability of the clinical, pathological, and biochemical features of Wilson’s disease explainable as an effect of changing environment or are we dealing with a genetically heterogeneous condition?” (B3, B5).Conceivably what we now recognize as Wilson’s disease is due to the abnormality of different alleles which may or may not be a t the same locus. It is also possible that the variability of the manifestations of the disease is due to one or several modifying genes. While these possibilities cannot at present be excluded, it seems equally possible that the biochemical features studied to date are not the primary products of a single abnormal gene, and the inconstancy of these secondary manifestations need not be interpreted as evidence of genetic heterogeneity. This latter possibility was suggested above. The answer will perhaps be given by more detailed metabolic and genetic studies on large families with Wilson’s disease, and by further inquiry into the nature of the basic abnormality in this condition. The latter will be possible only by increasing our existing meager knowledge of the intricate mechanisms whereby copper is handled by the human and animal organism. REFERENCES Al. Adelstein, S.J., and Vallee, B. L., Copper. In “Mineral Metabolism” (C. L. Comar and F. Bronner, eds.), Vol. 2, Part 2B, pp. 371-401. Academic Press, New York, 1962. A2. Adelatein, S.J., and Vallee, B. L., Copper metabolism in man. New Engl. J . Med. 266, 892-897 (1961). A3. Adelstein, 8. J., Coombs, T. L., and Vallee, B. L., Metalloenzymes and myocardial
COPPER METABOLISM
51
infarction. I. The relation between serum copper and ceruloplasmin and its catalytic activity. New Engl. J . Med. 266, 105-109 (1956). A4. Aisen, P., Schorr, J. B., Morell, A. G., Gold, R. Z., and Scheinberg, I. H., Rapid screening test for deficiency of plasma ceruloplasmin and its value in diagnosis of Wilson’s disease. Am. J . Med. 28, 550-554 (1960). A5. Akerfeldt, S., Oxidation of N,N-dimethyl-p-phenylenediamineby serum from patients with mental disease. Science 126, 117-119 (1957). A6. Alexander, G. V., Determination of zinc, copper, and iron in biological tissues-an X-ray fluorescence method. Anal. Chem. 34, 951-953 (1962). A7. Allan, J. E., The determination of copper by atomic absorption spectrophotometry Spectrochim. Acta 17, 459466 (1961). A8. Allen, W. D., and Schneiden, H., Effect of malnutrition on liver copper in the rat. Proc. SOC.Exptl. Biol. Med. 110, 346-347 (1962). A9. Ambe, K. S., and Venkataraman, A., Depolymerization of cytochrome oxidase to a water-soluble monomeric protein. Biochem. Biophys. Res. Commun. 1, 133-137 (1959). A10. Anderson, P. J., and Popper, H., Changes in hepatic structure in Wilson’s disease. Am. J . Pathol. 36, 483497 (1960). A l l . Angel, C., Leach, B. E., Martens, S., Cohen, M., and Heath, R. G., Serum oxidation tests in schizophrenic and normal subjects. A.M.A. Arch. Neurol. Psychiat. 78, 500-504 (1957). A12. Arima, M., and Kurumada, T., Genetical studies of Wilson’s disease in childhood. 11. Mode of inheritance and gene frequency in Japan. Paediat. Univ. Tokyo 7 , 7-12 (1962). A13. Axtrup, S., “The Blood Copper in Anaemias of Children with Special Reference to Premature Cases.” P. H. Lunstedt Univ. Bokhandel, Lund, 1946. B1. Bearn, A. G., Genetic and biochemical aspects of Wilson’s disease. Am. J . Med. 16, 442-449 (1953). B2. Bearn, A. G., Genetic aspects of Wilson’s disease. Proc. Roy. SOC.Med. 62, 60-62 (1959). B3. Bearn, A. G., A genetical analysis of thirty families with Wilson’s disease. Ann. Human Genet. 24, 3 3 4 3 (1960). B4. Bearn, A. G., The place of BAL in the therapy of Wilson’s disease. Am. J . Med. 21, 134 (1956). B5. Bearn, A. G., Wilson’s disease. In “The Metabolic Bask of Inherited Disease” (J. B. Stanbury, J. B. Wyngaarden, and D. S. Frederickson, eds.), pp. 809-838. McGraw-Hill, New York, 1960. B6. Beam, A. G., and Kunkel, H. G., Abnormalities of copper metabolism in Q‘ilson’s disease and their relationship to the aminoaciduria. J . C h . Invest. 33, 400-409 (1954). B7. Bearn, A. G., and Kunkel, H. G., Biochemical abnormalities in Wilson’s disease. J . Clin. Invest. 31, 616 (1952). B8. Bearn, A. G., and Kunkel, H. G., Localization of Cue4 in serum fractions following oral administration: An alteration in Wilson’s disease. Proc. Soe. Exptl. Biol. Med. 86, 44-48 (1954). B9. Beam, A. G., and Kunkel, H. G., Metabolic studies in Wilson’s disease using Cu“. J . Lab. Clin. Med. 46, 623-631 (1955). BlO. Bearn, A. G., and Kunkel, H. G., Wilson’s disease. Ergeb. I n n . Med. Kinderheilk. [N.S.] 7, 147-169 (1957).
52
ANDREW SASS-KORTSAK
B11. Bearn, A. G., and McKusick, V. A., Azure lunulae. An unusual change in the fingernails in two patients with hepatolenticular degeneration (Wilson’s disease). J. Am. Med. Assoc. 166, 904-905 (1958). B12. Bearn, A. G., Yu, T. F., and Gutman, A. B., Renal function in Wilson’s disease. J. CZin. Invest. 36, 1107-1114 (1957). B13. Behl, P. N., Agarwal, R. S., and Singh, G., The role of copper in vitiligo. J. Indian Med. Assoc. 37, 593-597 (1961). B14. Bennetts, H. W., and Chapman, F. E., Copper deficiency in sheep in Western Australia: A preliminary account of the etiology of enzootic ataxia in lambs and an anaemia of ewes. Australian vet. J. 13, 138-149 (1937). B15. Berman, E., An application of atomic absorption spectrometry in clinical chemistry: Determination of copper in biological material. Clin. Chem. 9,459460 (1963), abstract. B16. Blumberg, W. E., and Eisinger, J., Physical and chemical studies on ceruloplasmin. I. The relation between blue color and the valence states of copper. J. Biol. Chem. 238, 1675-1682 (1963). B17. Borchard, LeR. G., and Butler, J. P., Determination of trace amounts of copper. Application of the Bathocuproine reagent to pulp, paper and pulping liquors. Anal. Chem. 29, 414-419 (1957). B18. Bouchilloux, S., McMahill, P., and Mason, H. S., The multiple forms of mushroom tyrosinase. Purification and molecular properties of the enzymes. J. BWZ. Chem. 238, 1699-1707 (1963). B19. Boudin, G., and PBpin, B., “DBgBnBrescence h6patolenticulaire.” Masson, Paris, 1959. B20. Bowen, H. J., The determination of copper and zinc in biological material by activation analysis. Intern. J. Appl. Radialion Isotopes 4, 214-220 (1959). B21. Bowland, J. P., Braude, R., Chamberlain, A. G., Glascock, R. F., and Mitchell, K. G., The absorption, distribution and excretion of labelled copper in young pigs given different quantities, as sulphate or sulphide, orally or intravenously. Brit. J. Nutr. 16, 59-72 (1961). B22. Brand, T., and Takats, I., HistochemischeUntersuchungdes Kayser-Fleischerschen Hornhautringes. Graefes Arch. Ophthalmol. 161, 391-394 (1951). B23. Brendstrup, P., Serum copper, serum iron and total iron-binding capacity of serum in acute and chronic infections. Acta Med. Scnnd. 146, 315-325 (1953). B24. Brenner, W. Beitrage zur Kenntnis des Eisen- und Kupferstoffwechselsim Kindersalter: Serumeisen und Serumkupfer bei acuten und chronischen Infektionen. 2.Kin.derheik 66, 14-35 (1949). B25. Brenner, W., and Breier, A., Beitrage zur Kenntnis des Eisen- und Kupferstoffwechsels im Kindesalter: Zugleich ein Beitrag BUT Kenntnis der kindlichen Schirophrenie. 2.Kinderheilk. 66, 620-646 (1949). B26. Bridgman, O., and Smyth, F. S., Progessive lenticular degeneration. J. Nervous Mental Disease 99, 534-543 (1944). €327. Broman, L., Chromatographic and magnetic studies on human ceruloplasmin. Acta SOC.Med. Upsalien. 69, Suppl. 7, 1-85 (1964). B28. Broman, L., Separation and characterization of two ceruloplasmins from human serum. Nature 182, 1655-1657 (1958). B29. Broman, L., and Kjellin, K., A rapid semi-continuous method for purification of ceruloplasmin from human serum. Biochim. Biophys. A d a 83, 101-109 (1964). B30. Broman, L., Malmstrom, B. G., Aasa, R., and Vannghrd, T., Quantitative elec-
COPPER METABOLISM
53
tron spin resonance studies on native and denatured ceruloplasmin and laccase. J . MoZ. B i d . 6, 301-310 (1962). B31. Buchholtz, C. F., Chemische Untersuchung der Vanillenschoten (Siliqua vanillae). Repertorium Pharm. 2, 253 (1816). B32. Bush, J. A., Mahoney, J. P., Markowitz, H., Gubler, C. J., Cartwright, G. E., and Wintrobe, M. M., Studies on copper metabolism. XVI. Radioactive copper studies in normal subjects and in patients with hepatolent.icular degeneration. J . Clin. Invest. 34, 1766-1778 (1955). B33. Butler, E. J., and Newman, G. E., The urinary excretion of copper and its concentration in the blood of normal human adults. J . Clin. Pathol. 9, 157-161 (1956). B34. Butt, E. M., Nusbaum, R. E., Gilmour, T. C., and DiDio, S. L., Use of emission spectrograph for study of inorganic elements in human tissues. Am. J. Clin. Pathol. 24, 385-394 (1954). B35. Butt, E. M., Nusbaum, R. E., Gilmour, T. C., and DiDio, S. L.,Trace metal patterns in disease states. Am. J. Clin. Pathol. SO, 479-497 (1958). B36. Butterworth, C. E., Jr., Gubler, C. J., Cartwright, G. E., and Wintrobe, M. M., Studies on copper metabolism. XXVI. Plasma copper in patients with tropical sprue. PTOC.SOC.Ezptl. Bwl. Med. 98, 594-597 (1958). C1. Cartwright, G. E., The relationship of copper, cobalt,,and other trace elements to hemopoiesis. Am. J. Clin. Nutr. 3, 11-19 (1955). C2. Cartwright, G. E., and Wintrobe, M. M., Copper metabolism in normal subjects. Am. J . Clin. Nutr. 14, 224-237 (1964). C3. Cartwright, G. E., Gubler, C. J., Bush, J. A., and Wintrobe, M. M., Studies on copper metabolism. XI. Copper and iron metabolism in the nephrotic syndrome. J. Clin. Invest. 33, 665-698 (1954). C4. Cartwright, G. E., Markowita, H., Shields, G. S., and Wintrobe, M. M., Studies on copper metabolism. XXIX. A critical analysis of serum copper and ceruloplasmin concentrations in normal subjects, patients with Wilson’s disease and relatives of patients with Wilson’s disease. Am. J. Med. 28, 555-563 (1960). C5. Cartwright, G. E., Hodges, R. E., Gubler, C. J., Mahoney, J. P., D a m , K., Wintrobe, M. M., and Bean, W. B., Studies on copper metabolism. XIII. Hepatolenticular degeneration. J. Clin. Invest. 33, 1487-1501 (1954). C6. Cartwright, G. E., Hughley, C. M., Jr., Ashenbrucker, H., Fay, J., and Wintrobe, M. M., Studies on free erythrocyte protoporphyrin, plasma iron and plasma copper in normal and anemic subjects. Blood 3, 501-525 (1948). C7. Cassirer, R., Ein Fall von progressiver Linsenkernerkrankung. Neurol. Zentr. 82, 1284-1297 (1913). C8. Chalmers, T. C., Iber, F. L. and Uzman, L. L. Hepatolenticular degeneration (Wilson’s disease) as a form of idiopathic cirrhosis. New Engl. J. Med. 266, 235-242 (1957). C9. Chou, T. P., and Adolph, W. H., Copper metabolism in man. Biochem. J. 29, 476-479 (1935). C10. Church, A. H., Researches on turacin, an animal pigment containing copper. Phil. Trans. Roy. Soc. 169, 627-636 (1869). C11. Colombo, J. P., and Richterich, It., Zur Bestimmung des Ckuloplasmins im Plasma. Schweiz. Med. Wochachr. 94, 715-720 (1964). C12. Cooper, A. M., Eckhardt, R. D., Faloon, W. W., and Davidson, C. S., Investigation of the aminoaciduria in Wilson’s disease (hepatolenticular degeneration): Demonstration of a defect in renal function. J. Clin. Invest. 29, 265-278 (1960).
54
ANDREW SASS-KORTSAK
C13. Cox, D., and Sass-Kortsak, A., Unpublished data (1964). C14. Cox, D. H., and Hale, 0. M., Dietary hormones and fat and serum cholesterol, transaminases and copper in swine. J . Nutr. 72, 77-80 (1959). C15. Crampton, R. F., Matthews, D. M., and Poisner, R. The absorption of copper by sacs of hamster small intestine. J. Physiol. (London)170, 27p-28p (1963). C16. Cumings, J. N., L‘HeavyMetals and the Brain.’’ Thomas, Springfield, Illinois, 1959. C17. Curzon, G., Effect of chymotrypsin on ceruloplasmin. Nature 181, 115-116 (1958). C18. Curzon, G., The effects of some ions and chelating agents on the oxydase activity of caeruloplasmin. Biochem. J. 77, 66-73 (1960). C19. Curzon, G., Reversible decolorization of caeruloplasmin under acid conditions. Biochim. Bwphya. Acta 71, 249-250 (1963). C20. Curzon, G., and Vallet, L., The purification of human ceruloplasmin. Biochem. J . 74, 279-287 (1960). D1. DeJorge, F. B., Canelas, H. M., Dim, J. C., and Cury, L., Studies on copper metabolism. 111. Copper contents of saliva of normal subjects and salivary glands and pancreas on autopsy material. Clin. Chim. Acta 9, 148-150 (1964). D2. Denny-Brown, D., “Diseases of the Basal Ganglia and Subthalamic Nuclei,” Vol. 6. Oxford Univ. Press, London and New York, 1945. D3. Denny-Brown, D., and Porter, H., The effect of BAL (2,3dimercaptopropanol) on hepatolenticular degeneration (Wilson’s disease). New Engl. J. Med. 246, 917-925 (1951). D4. Deutsch, H. F., Kaaper, C. B., and Walsh, D. A., Rapid method for preparation of crystalline human ceruloplasmin from Cohn fraction IV-1. Arch. Bwchem. BiOphy8. 99, 132-135 (1962). D5. Dick, A. T., Molybdenum in animal nutrition. Soil Sci. 81, 229-236 (1956). D6. Dick, A. T., Studies on the assimilation and storage of copper in crossbred sheep. Australian J. Agr. Res. 6, 511-544 (1954). El. Earl, C. J., Anatomical distribution of copper in human brain. I n “Wilson’s Disease, Some Current Concepts” (J. M. Walshe and J. N. Cumings, eds.), pp. 18-23. Thomaa, Springfield, Illinois, 1961. E2. Earl, C. J., Moulton, M. J., and Selverstone, B., Metabolism of copper in Wilson’s disease and in normal subjects: Studies with Cu”. Am. J. Med. 17, 205-213 (1954). E3. Eden, A., and Green, H. H., Micro-determination of copper in biological material. Biochem. J . 34, 1202-1208 (1940). E4. Edozien, J. C., and Udeozo, I. O., Serum copper, iron and iron binding capacity in Kwashiorkor. J. Trop. Pediat. 8, 60-64 (1960). E5. El-Mofty, A., El-Mofty, A. M., Abdelal, H., and El-Hawary, M. F., Studies on mode of action of psoralen derivatives. 11. Pituitary-adrenal axis control of copper metabolism and its response to psoralens. J. Invest. Dermatol. 32, 651-659 (1959). E6. Elsner, P., and Hornykiewitz, O., Die Beeinflussbarkeit der p-Polyphenoloxydaseactivitat des menschlichen Blutserums durch Sexualhormone. Arch. U y w k o l . 186, 251-257 (1954). E7. Elvehjem, C. A., The biological significance of copper and its relation to iron metabolism. Physiol. Rev. 16, 471-507 (1935). EEL Enger, E., Wilson’s disease, report of a case with normal serum ceruloplaamin level. Acta Med. Scand. 163, 121-124 (1959). F1. Fanshier, D. W., and Kun, E., Properties of 8-mercaptopyruvate transsulfurase
COPPER METABOLISM
55
chromatographically resolved in the presence of pmercaptoethanol. Biochim. Biophys. Acta 68, 266-278 (1962). F2. Fazekas, I. Gy., Romhanyi, I., and Rengei, B., Copper content of fetal organs (in Hungarian). Kiserl. Orvostud. 16,23Ck238 (1963). F3. Finby, N., and Bearn, A. G., Roentgenographic abnormalities of the skeletal system in Wilson’s disease (hepatolenticular degeneration). Am. J. Roenlgenol. Radium Therapy Nucl. Med. 79, 603-611 (1958). F4. Fleischer, B., Zwei weitere Falle von griinlicher Verfiirbung der Kornea. Klin. Monatsbl. Augenheilk. 41, 489-491 (1903). F5. Fleischer, B., Uber eine der “Pseudosklerose” nahestehende bisher unbekannte Krankheit (gekennzeichnet durch Tremor, psychische Storungen, briiunliche Pigmentierung bestimmter Gewebe, insbesondere auch der Hornhautperipherie, Leberzirrhose). Deul. Z . Nercenheilk. 44, 179-201 (1912). 1’6. Fling, M., Horowitz, M. H., and Heinemann, S. F., The isolation and properties of cryst,allinetyrosinase from neurospora. J. Biol. Chem. 238, 2045-2053 (1963). F7. Fontaine, M., and Leloup, J., Action antit,hyroidiens (aminothiazole et ph6nylthiourea) sur la cuprhmie des verthbrbs. poicilothermes. Compt. R e d . Sac. Bwl. 141, 148 (1947). F8. Fredericq, L., Recherche8 sur la physiologie du poulpe commun (Octopus vulgaris). Arch. Zool. Expll. Gen.7, 535-583 (1878). F9. Fritze, K., Preparation of high specific activity CUM. Radiochim. Acia 3, 166-167 (1964). F10. Fritze, K., Aspin, N., and Holmes, T. H., Routine determination of copper in biological materials by neutron activation. Radwchim. Acla 8, 204-206 (1964). F l l . Frohman, C. E., Goodman, M. Luby, E. D., Beckett, P. G. S., and Senf, R., Ceruloplasmin, transferrin, and tryptophan in schizophrenia. A.M.A. Arch. Neurol. Psychid. 79, 730-734 (1958). GI. Gerlach, W., Untersuchungen uber den Kupfergehalt menschlicher und tierischer Organe. Virchow’s Arch. Palhol. Anat. Physwl. 294, 171-197 (1934). G2. Gerlach, W., Uber den Kupfergehalt menschlicher Organe in beaonderen Fallen. Virchow’s Arch. Pathol. A d . Physiol. 296, 394-398 (1935). G3. Gerlach, W., and Rohrschneider, W., Besteht das Pigment dea Ka.yser-Fleischerschen Hornhautringes aus Silber? Klin. Wochschr. 13, 48-49 (1934). G4. German, J. L., 111, and Bearn, A. G., Effect of estrogens on copper metabolism in Wilson’s disease. J. Clin. Imest. 40, 445453 (1961). G5. Giorgio, A. J., Cartwright, G. E., and Wintrobe, M. M., Determination of urinary copper by means of direct extraction with zinc dibenzyl dithiocarbamate. Am. J. Clin. Pathol. 41, 22-26 (1964). G6. Gitlin, D., Hughes, W. L., and Janeway, Ch. A., Absorption and excretion of copper in mice. Nature 188, 150-151 (1960). G7. Glazebrook, A. J., Wilson’s disease. Edinburgh Med. J. 62, 83-87 (1945). G8. Goldfischer, S., Liver cell lysosomes in Wilson’s disease: acid phosphatase activity by light and electron microscopy. Am. J. Pathol. 43, 511-518 (1963). G9. Gopalan, C., Reddy, V., and Mohan, V. S., Some aspects of copper metabolism in protein-calorie malnutrition. J. Pediat. 63, 646-649 (1963). G10. Gran, G., The use of oxalyldihydrazidein a new reaction for the spectrophotometric microdetermination of copper. A d . Chim. Acta 14, 150-152 (1956). G11. Gruter, W., Hiimolytische Krisen als Fruhmanifestation der Wilaonschen Krank-
56
ANDREW SASS-KORTSAK
heit. Ein Beitrag zur Pathogenese der hepatocerebralen Degeneration. Deut. 2. Nervenheilk. 179, 401-422 (1959). G12. Gubler, C. J., Lahey, M. E., Cartwright, G. E., and Wintrobe, M. M., Studies on copper metabolism. IX. The transportation of copper in blood. J. Clin. Invest. S2, 405-414 (1953). G13. Gubler, C. J., Brown, H., Markowita, H., Cartwright, G. E., and Wintrobe, M. M., Studies on copper metabolism. XXIII. Portal (Laennec’s)cirrhosis of liver. J. Clin. Invest. 36, 1208-1216 (1957). G14. Gubler, C. J., Lahey, M. E., Ashenbrucker, H., Cartwright, G. E., and Wintrobe, M. M., Studies on copper metabolism. I. Method for the determination of copper in whole blood, red blood cells, and plasma. J . BWZ. Chem. 196,209-220 (1952). G15. Guerithault, M. B., Sur la prbence du cuivre dam les plantes et particulibrement dans les matibres alimentaires d’origine vegbtale. Compt. Rend. Acud. Sei. 171, 196-198 (1920). G16. Guggenheim, K., The role of zinc, copper, and calcium in the etiology of “meat anaemia.” Blood 2S, 786-794 (1964). G17. Gupta, J. C., Rangam, C. M., and Bhagwat, A. G., Effect of acute liver injury on serum iron and copper in guinea pigs. Indim J. Med. Sci. 16, 504-507 (1962). H1. Hall, H. C., “La d6g6drescence hkpato-lenticulaire; Maladie de Wilson, pseudosclbose.” Masson, Paris, 1921. H2. Hansl, N., Hyman, L., Luparello, T., and Zimdahl, W. T., Studies on hepatolenticular degeneration. J . Lab. Clin. Med. 48, 108-116 (1956). H3. Harris, R. S., and Scheinberg, I. H., Hepatic function in a child lacking ceruloplasmin. Gastroenterology 34, 1049-1050 (1958). H4. Hart, E. B., Steenbock, H., Waddell, J., and Elvehjem, C. A., Iron in nutrition. VII. Copper as a supplement to iron for hemoglobin building in the rat. J. B i d . Chem. 77, 797-812 (1928). H5. Hawke, W. M., Unpublished observations (1964). H6. Hawke, W. M., and Holmes, T. H., Unpublished observations (1964). H7. Heijkenskjold, F., and Hedemstedt, S., Serum copper determinations in normal pregnancy and abortion. Acfa Obstet. Gynecol. Scand. 41,4147 (1962). H8. Heilmeyer, L., Kinderling, W., and Stuwe, G., “Kupfer und Eisen als korpereigene Wirkstoffe und ihre Bedeutung beim Krankheitsgeschehen.” Fischer, Jena, 1941. H9. Hejduk, J., Untersuchungen uber das Verhalten von Eisen und Kupfer im Blutserum der Frauen in der Schwangerschaft, unter der Geburt, im Wochenbett sowie bei Neugeborenen. 2. Geburtshilfe Cynaelcol. 160, 187-199 (1963). H10. Hermann, G. E., and Kun, E., Intracellular distribution of copper in rat liver and its response to hypophysectomy and growth hormone. Exptl. Cell Res. 22, 257-263 (1961). H11. Hitzig, W. H., Das Bluteiweissbild beim gesunden Saugling. Speaifische Proteinbestimmungen mit besonderer Berucksichtigung immunochemischer Methoden. Helv. Paediat. A d a 16,46-81 (1961). H12. Hodges, R. E., Kirkendall, W. M., and Gubler, C. J., Some aspecta of kidney function in hepatolenticular degeneration (Wilson’s disease). J . Lab. Clin. Med. 47, 337-342 (1956). H13. Holmberg, C. G., and Laurell, C. B., Investigations in serum copper. I. Nature of serum copper and its relation to the iron-binding protein in human serum. Actu Chem. Scand. 1, 944-950 (1947). H14. Holmberg, C. G., and Laurell, C. B., Investigations in serum copper. 11. Isolation
COPPER METABOLISM
57
of the copper-containing protein and the description of some of its properties. Acta Chem. Scand. 2, 550-556 (1948). H15. Holmberg, C. G., and Laurell, C. B., Oxidase reactions in human plasma caused by coeruloplasmin. Scand. J . Clin. Lab. Invest. 3, 103-107 (1951). H16. Holmberg, C. G., and Laurell, C. B., Investigations in serum copper. IV. Effect of different anions on the enzymatic activity of coeruloplasmin. Acta Chem. Scand. 6,921-930 (1951). H17. Horwitt, M. K., Meyer, B. J., Meyer, A. C., Harvey, C. C., and Haffron, D., Serum copper and oxidaae activity in schizophrenic patients. Correlations with erythrocyte sedimentation rate, C-reactive protein, ascorbic acid, basal metabolic rate, and sulfobromophthalein (bromsulphalein) retention tests. A.M.A. Arch. Neurol. Psychiat. 78, 275-282 (1957). H18. Hoste, J., On a new copper specific group. Anal. Chim. Acta 4, 23-37 (1950). H19. Howel, J. S., Histochemical demonstration of copper in copper fed rats and in hepatolenticular degeneration. J . Pathol. Bacterwl. 77, 473484 (1959). H20. Humoller, F. L., Majka, F. A., Barak, A. J., Stevens, J. D., and Holthaus, J. M., Determination of plasma amine oxidase activity. Clin. Chem. 4, 1-12 (1958). H21. Hunt, A. H., Parr, R. M., Taylor, D. M., and Trott, N. G., Relation between cirrhosis and trace metal content of liver with special reference to primary biliary cirrhosis and copper. Brit. Med. J . 11, 1498-1501 (1963). 11. Iber, F. L., Chalmers, T. C., and Uzman, L. L., Studies of protein metabolism in hepatolenticular degeneration. Metab. Clin. Exptl. 6, 388-396 (1957). J1. Jensen, W. N., and Kamin, H., Copper transport and excretion in normal subjects and in patients with Laennec’s cirrhosis and Wilson’s disease: A study with CUM. J . Lab. Clin. Med. 49, 200-210 (1957). 52. Jeunet, F., Richterich, R., and Aebi, H., Bile et Chloplasmine. Etude in d r o A l’aide de la perfusion du foie isolb. J. Physiol. (Paris)64, 729-737 (1962). 53. Johnson, N. C., Electrometric titration of copper and zinc in biological material. Clin. Chem. 8, 497-501 (1962). 54. Johnson, N. C., Kheim, T., and Kountz, W. B., Influence of sex hormones on total serum copper. Proc. SOC.Exptl. Biol. Med. 102,98-99 (1959). 55. Jones, P. D., and Newman, E. J., The determination of copper in some food stuffs Analyst 87,637-642 (1962). with 2,9dirnethyl-lJl0-phenanthroline. K1. Kaiser, D. G., and Meinke, W. W., Rapid activation analysis of trace copper in tissue using 5.1-minute copper-66. Anal. Biochem. 6, 77-81 (1963). K2. Kanabrocki, E. L., Fields, T., Decker, C. F., Case, L. F., Miller, E. B., Kaplan, E., and Oester, Y. T., Neutron activation studies of biological fluids: Manganese and copper. Intern. J. Appl. Radiation Isotopes 16, 175-190 (1964). K3. Kasper, C. B., and Deutsch, H. F., Physicochemical studies of human ceruloplasmin. J. Bwl.Chem. 238, 2325-2337 (1963). K4. Kasper, C. B., Deutsch, H. F., and Beinert, H., Studies on the state of copper in native and modified human ceruloplasmin. J. Bwl. Chem. 238, 2338-2342 (1963). K5. Kautzsch, E., Eisen- und Kupferstoffwechselbei malignen Neoplasien. Med. Klin. 40, 1851-1858 (1959). K6. Kayser, B., Vber einen Fall von angeborener griinlicher Verfiirbung der Cornea. Klin. Monatsbl. Augenheilk. 40, 22-25 (1902). K7. Kimmel, J. R., Markowitz, H., and Brown, D. M., Some chemical and physical propertiea of erythrocuprein. J. Bwl. Chem. 234, 46-50 (1959).
58
ANDREW SASS-KORTSAK
K8. Kjellii, K., Determination of copper in cerebrospinal fluid by activation analysis. J . Neurochem. 10, 89-93 (1963). K9. Kleinbaum, H., Kupferstoffwechselbilanren bei Sauglingen. 2. Kinderheilk. 87, 101-115 (1962). K10. Kleinbaum, H., Voriibergehende Bypoproteiniimie mit Hypocupramie und Eisenmangelanamie bei dystrophen Sauglingen. 2.Kinderheilk. 88, 27-34 (1963). K11. Kleinmann, H., and Klinke, J., tfber den Kupfergehalt menschlicher Organe. Virchow’s Arch. Pathol. Anat. Physwl. 276, 422-435 (1929). K12. Knehr, C. A., and Bearn, A. G., Psychological impairment in Wilson’s disease. J. Nervous Mental Disease 124, 251-255 (1956). K13. Koch, H. J., Jr., Smith, E. R., Shimp, N. F., and Connor, J., Analysis of trace elements in human tissues. I. Normal tissues. Cancer 9, 499-511 (1956). K14. Koch, H. J., Jr., Smith, E. R., and McNeely, J., Analysis of trace elements in human tissues. 11. Lymphomoatous diseases. Cancer 10, 151-160 (1957). K15. Krebs, H. A., uber das Kupfer im menschlichen Blutserum. Klin. Wochschr. 7 , 584-585 (1928). L1. Lahey, M. E., Iron and copper in infant nutrition. Am. J. Clin. Nutr. 6, 516-526 (1957). L2. Lahey, M. E., Behar, M., Viteri, F., and Scrimshaw, N. S., Values of copper, iron and iron binding capacity in serum in kwashiorkor. Pediatrics 22, 72-79 (1958). L3. Lahey, M. E., Gubler, C. J., Chase, M. S., Cartwright, G. E., and Wintrobe, M. M., Studies on copper metabolism. VII. Blood copper in pregnancy and various pathological states. J . Clin. Invest. 32, 329-339 (1953). L4. Laurell, C. B., Metal-binding plasma proteins and cation transport. 1%“The Plasma Proteins” (F. W. Putnam, ed.), Vol. 1, pp. 349-378. Academic Press, New York, 1960. L5. Lawrie, N. R., and Carter, R. A., Acute case of Wilson’s disease (hepatolenticular degeneration). Lancet I, 1309-1311 (1958). L6. Leverton, R. M., and Binkley, E. S., Copper metabolism and requirement of young women, J . Nutr. 27, 43-53 (1944). L7. Levine, W. G., and Peisach, J., Ethylenediaminetetraacetate,iron and ceruloplasmin activity. Bwchim. Bwphys. Acta 77, 602-614 (1963). L8. Liebhafsky, H. A,, and Winslow, E. H., Diphenylthiocarbazone (Dithirone) as an analytical reagent. J. Am. Chern. Soe. 69, 1966-1971 (1937). L9. Long, M. J., Metal ion analysis by emission spectroscopy of starch block electrophoretically separated human serum proteins. Am. J. Med. Technol. 28, 241-256 (1962). M1. McAlister, R., Martin, G. M., and Beneditt, E. P., Evidence for multiple caeruloplasmin components in human serum. Nature 190, 927-929 (1961). M2. MacDonald, G. M., The copper content of some New Zealand autopsy livers with the reference to the copper content of associated water supplies. New Zealand Med. J . 62, 620-622 (1963). M3. Macdonald, I., and Warren, P. J., The copper content of the liver and hair of African children with Kwaahiorkor. Brit. J. Nutr. 16, 593-596 (1961). M4. McFarlane, W. D., Application of the sodium diethyldithiocarbamate reaction to . the micro-colorimetric determination of copper in organic substances. Bhchern. J. 96, 1022-1033 (1932). M5. Mahler, H. R., Uricase. In “The Enzymes” (P. D. Boyer, H, Lardy, and K. Myrback, eds.), Vol. 8, Part B, pp. 285-296. Academic Press, New York, 1963.
COPPER METABOLISM
59
M6. Mahoney, J. P., Bush, J. A., Gubler, C. J., Moretz, W. H., Cartwright, G. E., and Wintrobe, M. M., Studies on copper metabolism. XV. Excretion of copper by animals. J. Lab. CZin. Med. 46, 702-708 (1955). M7. Malmstrom, B. G., and Neilands, J. B., Metalloproteins. Ann. Rev. Biochem. 33, 331-354 (1964). M8. Malmstrom, B. G., Broman, L., and Mosbach, R., A magnetic susceptibility study of copper valence in ceruloplasmin and laccase. J. Mol. Bwl.6,450452 (1962). M9. Mandelbrote, B. M., Stanier, M. W., Thompson, R. H. S., and Thurston, M. N., Studies on copper metabolism in demyelinating diseases of the central nervous system. Brain 71, 212-228 (1948). M10. Mann, T., and Keilin, D., Haemocuprein and hepatocuprein, copper-protein compounds of blood and liver of mammals. Proc. Roy. SOC.126, 303-315 (1938). M11. Markowitz, H., Cartwright, G. E., and Wintrobe, M. M., Studies on copper metabolism. XXVII. The isolation and properties of an erythrocyte cuproprotein (Erythrocuprein). J. Biol. Chem. 234, 40-45 (1959). M12. Markowitz, H., Gubler, C. J., Mahoney, J. P., Cartwright, G. E., and Wintrobe, M. M., Studies on copper metabolism. XIV. Copper, ceruloplasmin and oxidase activity in sera of normal human subjects, pregnant women, and patients with infection, hepatolenticular degeneration and nephrotic syndrome. J. Clin. Invest. 84, 1498-1508 (1955). M13. Martens, R. I., and Githem, R. E., Sr., Small amounts of copper in dyes and rubber chemicals. Anal. Chem. 24, 991-993 (1952). M14. Martin, G. M., Copper content of hair and nails of normal individuals and of patients with hepatolenticular degeneration. Nature 202, 903-904 (1964). M15. Matthews, W. B., The absorption and excretion of radio-copper in hepatolenticular degeneration (Wilson’s disease). J. Neurol. Neurosurg. Psychiat. 17, 242-246 (1954). M16. Matthews, W. B., Milne, M. D., and Bell, M., The metabolic disorder in hepatolenticular degeneration. Quart. J. Med. 21, 425-446 (1952). M17. Maytum, W. J., Goldstein, N. P., McGuckin, W. F., and Owen, C. A., Jr., Copper metabolism in Wilson’s disease, Laennec’s cirrhosis and hemachromatosis: Studies with radiocopper (CUM). Proc. Sta$ Meetings Mayo Clinic 36, 641-660 (1961). M18. Meissner, W., Sur la pr6aence du cuivre dans les cendres vegbtaux. Ann. Chim. Phys. 4, 106 (1817). M19. Meyer, B. J., Meyer, A. C., and Horwitt, M. K., Effect of triiodothyronine on serum copper and basal metabolism in schizophrenic patients. Arch. Gen. Psychiat. 1, 372-378 (1959). M20. Meyer, B. J., Meyer, A. C., and Horwitt, M. K., Factors influencing serum copper and ceruloplasmin oxidative activity in the rat. Am. J. Physwl. 194,581-584 (1958). M21. Mills, C. F., Copper complexes in grassland herbage. Biochem. J . 67, 603-610 (1954). M22. Mills, C. F., Availability of copper in freezedried herbage and herbage extracts to copperdeficient rats. Brit. J. Nutr. 9, 398-409 (1955). M23. Mills, C. F., Studies on the copper compounds in aqueous extracts of herbage. Bwchem. J. 63, 187-190 (1956). M24. Mills, C. F., The dietary availability of copper in the form of naturally occurring organic complexes. Biochem. J. 63, 190-193 (1956). M25. Mills, C. F., Metabolic interrelationships in the utilization of trace elements. Proc. Nutr. SOC.Engl. Scot. Z3, 38-45 (1964).
60
ANDREW SASS-KORTSAK
M26. Mitchell, H. H., and Hamilton, T. S.,Dermal excretion under controlled environmental conditions of nitrogen and minerals in human subjects with particular reference to calcium and iron. J . Bwl. Chem. 178, 345-361 (1949). M27. Mohamed, M. S., and Greenberg, D. M., Isolation of purified copper protein from horse liver. J . Den. Physiol. 37, 433-439 (1954). M28. Monckeberg, F., Vildbola, J., Figueroa, M., Oxman, S., and Meneghello, J., Hematologic disturbances in infantile malnutrition. Values for copper, iron, paraphenylene diamine oxidase and iron-binding capacity in the serum. Am. J . Clin. Nutr. 11, 525-529 (1962). M29. Moore, I., Constable, B. J., Day, K. C., Impey, S. G., and Symonds, K. R., Copper deficiency in rats fed upon raw meat. Brit. J . Nutr. 18, 135-146 (1964). M30. Morell, A. G., and Scheinberg, I. H., Preparation of an apoprotein from ceruloplasmin by reversible dissociation of copper. Science 127, 588-590 (1958). M31. Morell, A. G., and Scheinberg, I. H., Heterogeneity of human ceruloplasmin. Science 131, 930-932 (1960). M32. Morell, A. G., Shapiro, J. R., and Scheinberg, I. H., Copper binding protein from human liver. I n “Wilson’s Disease, Some Current Concepts” (J. M. Walshe and J. N. Cumings, eds.), pp. 36-41. Thomas, Springfield, Illinois, 1961. M33. Morell, A. G., Aisen, P., Blumberg, W. E., and Scheinberg, I. H., Physical and chemical studies on ceruloplasmin. 11. Molecular oxygen and the blue color of ceruloplasmin. J . Biol. Chem. 239, 1042-1043 (1964). M34. Morgan, H. G.; Stewart, W. K., Lowe, K. G., Stowers, J. M., and Johnstone, J. H., Wilson’s disease and the Fanconi syndrome. Quart. J . Med. 31, 361-384 (1962). N1. National Academy of Sciences. National Research Council Publ. No. 598, Recommended Dietary Allowances,” p. 21. U.S. Govt. Printing Office, Washington, D.C., 1958. N2. Neal, W. M., Becker, R. B., and Shealy, A. L., A natural copper deficiency in cattle rations. Science 74, 418-419 (1931). N3. Neumann, P. Z., and Sass-Kortsak, A., Binding of copper by serum proteins. Voor Sanguinis 8, 111-112 (1963). N4. Neumann, P. Z., Carr, R. I., and Sass-Kortsak, A., The handling of a single intravenous dose of copper by the rat. Can. Med. Assoc. J . 68, 229 (1962). N5. Newrnan, G. E., and Ryan, M., The determination of copper in biological materials by flame spectrophotometry. J . Clin. Puthol. 16, 181-184 (1962). N6. Nishimura, H., Zur Histopathologie der Leberschadigung, insbesondere der Zirrhosen im Zusammenhang mit dem Gehalt der Leber an Kupfer und Eisen. Differenzierung anhand von histochemischen und chemisch-quantitativen Versuchen. Actu Pathol. Jupon. 12, 59-89 (1962). N7. Nusbaum, R. E., Alexander, G. J., Butt, E. M., Gilmour, T. C., DiDio, S. L., Some spectrographic studies of trace element storage in human tissues. Soil Sci. 86, 95-99 (1958). N8. Nusbaum, R. E., Butt, E. M., Gilmour, T. C., and DiDio, S. L., Analysis of human tissue ash with a direct reading spectrometer. Am. J . Clin. Pathol. 36, 44-52 (1961). 01. Osborn, S. B., and Walshe, J. M., Copper uptake by the liver: Study of a Wilson’s disease family. In “Wilson’s Disease, Some Current Concepts” (J. M. Walshe and J. N. Cumings, eds.), pp. 141-150. Thomas, Springfield, Illinois, 1961. 02. Osborn, S. B., and Walshe, J. M., Effects of penicillamine and dimercaprol on turnover of copper in patients with Wilson’s disease. Lancet I, 70-73 (1958).
COPPER METABOLISM
61
03. Osborn, S. B., and Walshe, J. M., Studies with radiocopper (CUM) in Wilson’s disease: The liver/thigh ratio. Clin. Sci. 27, 319-328 (1964). 04. Osborn, S. B., Roberts, C. N., and Walshe, J. M., Uptake of radiocopper by the liver. A study of patients with Wilson’s disease and various control groups. Clin. Sci. 24, 13-22 (1963). 05. Owen, C. A., Jr., Copper metabolism and Wilson’s disease. I n “Dynamic Clinical Studies with Radioisotopes” (R. M. Kniseley, W. N. Tame, and E. B. Anderson, eds.), pp. 581-606. U. S. Atomic Energy Commission, Division of Technical Information, Oak Ridge, Tenneesee, 1964. P1. Pagliardi, E., Giangrandi, E., and Vinti, A., Comportamento del rame plasmatico ed eritrocitario in condizioni morbose. Rass. Fiswpatol. Clin. Terap. 29, 907-921 (1957). P2. Pagliardi, E., Giangrandi, E., and Vinti, A., Erythrocyte copper in iron deficiency anaemia. Acta Haematol. 19,231-240 (1958). P3. Pedersen, K. O., In “Les proteins: rapports et discussions” (R. Soops, ed.). Univ. Bruxelles, Brueaels, 1953. P4. Peisach, J., and Levine, W. G., On the mechanism of ceruloplasmin-catalyzed oxidations. Bwchim. Biophys. Acta 77, 615-628 (1963). P5. Peterson, R. E., and Bollier, M. E., Spectrophotometric determination of serum copper with bis-cyclohexanoneoxalyldihydrazone.Anal. Chem. 27,1195-1197 (1955). P6. Pineda, E. P., Ravin, H. A., and Rutenburg, A. M., Serum ceruloplasmin: Observations in patients with cancer, obstructive jaundice and other diseases. Gastroenkrobgy 43, 266-270 (1962). P7. Plum, C. M., and Hansen, 5. E., Studies on variations in serum copper and serum copper oxidase activity together with studiea on the copper content of the cerebrospinal fluid, with particular reference to the variations in multiple sclerosis. Acta Psychiat. Neurol. Scad. 36, Suppl. 148, 41-78 (1960). P8. Pomerantz, S. H., Separation, purification, and properties of two tyrosinases from hamster melanoma. J. Biol. Chem. 238, 2351-2357 (1963). P9. Porter, H., Copper excretion in the urine of normal individuals and of patients with hepatolenticular degeneration (Wilson’s disease). Arch. Biochem. Biophys. 31, 262-265 (1951). P10. Porter, H., Copper-protein combinations in the brain in Wilson’s disease. In “Wilson’s Disease, Some Current Concepts” (J. M. Walshe and J. N. Cumings, eds.), pp. 24-35. Thomas, Springfield, Illinois, 1961. Pll. Porter, H., The intracellular distribution and chromatographic separation of copper proteins in Wilson’s disease. Trans. Am. Neurol. Assoc. 88, 159-164 (1963). P12. Porter, H., and Ainsworth, S., The isolation of the copper-containing protein cerebrocuprein I from normal human brain. J. Neurochem. 6,91-98 (1959). P13. Porter, H., Johnston, J., and Porter, E. M., Neonatal hepatic mitochondrocuprein. I. Isolation of a protein fraction containing more than 4% copper from mitochondria of immature bovine liver. Bwchim. Biophys. Acta 66, 66-73 (1962). P14. Porter, H., Sweeney, M., and Porter, E. M., Human hepatocuprein. Isolation of a copper-protein from the subcellular soluble fraction of adult human liver. Arch. Biochem. Biophys. 106, 319-325 (1964). P15. Porter, H., Sweeney, M., and Porter, E. M., Neonatal hepatic mitochondrocuprein. 11. Isolation of the copper-containing subfraction from mitochondria of newborn human liver. Arch. Bwchem. Biophys. 104, 97-101 (1964).
62
ANDREW SASS-KORTSAK
P16. Poulik, M. D., Electrophoretic and immunological studies on structural subunits of human ceruloplasmin. Nature 194, 842-844 (1962). P17. Poulik, M. D., Heterogeneity and structural subunits of human ceruloplasmin. I n “Protides of the Biological Fluids” (H. Peeters, ed.), Vol. 10, pp. 170-182. Elsevier, Amsterdam, 1963. P18. Poulik, M. D., and Bearn, A. G., Heterogeneity of ceruloplasmin. Clin. Chim. Actu 7, 374-382 (1962). R1. “Radiological Health Handbook.” Division of Radiological Health, U. S. Department of Health, Education and Welfare. Washington, D.C., 1960. R2. Rangam, C. M., and Bhagwat, A. G., Serum iron and copper levela in jaundice. Indian J . Med. Sci. 16, 499-503 (1962). R3. Ravensteyn, A. H. Van. See V4. R4. Ravin, H. A., An improved colorimetric enzymatic assay of ceruloplasmin. J . Lab. CZin. Med. 68, 161-168 (1961). R5. Rechenberger, J., Serumeisen und Serumkupfer bei akuten und chronischen Leukamien sowie by Morbus Hodgkin. Deut. Z . Verdauungs- Sbflwechselkrankh. 17, 78-85 (1957). R6. Rechenberger, J., Serumeisen und Serumkupfer bei der Hepatitis epidemica und den posthepatitischen Lebererkrankungen. Verhandl. Deut. Ges. Inn. Med. 63, 317-321 (1957). R7. Rees, K. R., Copper as an enzyme poison. I n “Wilson’s Disease, Some Current Concepts” (J. M. Walshe, and J. N. Cumings, eds.), pp. 49-51. Thomas, Springfield, Illinois, 1961. R8. Reiff, B., and Schneiden, H., Plasma copper and iron levels and plasma paraphenylene diamine oxidase activity (plasma copper oxidase activity) in kwashiorkor. Blood 14, 967-971 (1959). R9. Report of the International Commission on Radiological Protection Committee I1 on permissible dose for internal radiation (1959), with bibliography for biological, mathematical, and physical data. Health Phys. 3, 1-380 (1960). R10. Rice, E. W., Correlation between serum copper ceruloplasmin activity and Greactive protein. Clin. Chim. Acta 6, 632-636 (1960). Rll. Rice, E. W., Spectrophotometric determination of serum copper with oxalyldihydrazide. J . Lab. Clin. Med. 66, 325-328 (1960). R12. Rice, E. W., Standardization of ceruloplasmin activity in terms of international enzyme units. Ann. Bwchem. 8, 452 (1962). R13. Rice, E. W., and Goldstein, N. P., Copper content of hair and nails in Wilson’s disease (hepatolenticular degeneration). Metab. Clin. Exptl. 10, 1085-1087 (1961). R14. Rice, E. W., Olson, R. E., and Sweeney, P. D., A study of serum copper and certain “acute-phase reactants” in alcoholics. Quart. J. Studies A h h o l 2 2 , 544-549 (1961). R15. Richterich, R., Temperli, A., and Aebi, H., Die Heterogenitiit des Caeruloplasmins: Isolierung und Characterisierung von zwei Cupro-Proteinen aus Humanserum. Bwehim. Biophys. Actu 66, 240-251 (1962). R16. Richterich, R., Gautier, E., Stillhart, H., and Rossi, E., The heterogeneity of caeruloplasmin and the enzymatic defect in Wilson’s disease. Helv. Paedial. Actu 16, 4-36 (1960). R17. Rosenoer, V. M., In “Wilson’s Disease, Some Current Concepts” (J. M. Walshe and J. N. Cumings, eds.), pp. 11G112. Thomas, Springfield, Illinois, 1961. R18. Rosenoer, V. M., Bone changes in Wilson’s disease. I n “Wilson’s Disease, Some
COPPER METABOLISM
63
Current Concepts” (J. M. Walshe and J. N. Cumings, eds.), pp. 245-246. Thomas, Springfield, Illinois, 1961. R19. Rosenoer, V. M., and Franglen, G. T., Ceruloplasmin in Wilson’s disease. Lancet 1163-1164 (1959). R20. Rumpel, A., Uber das Wesen und die Bedeutung der Leberveranderungenund der Pigmentierungen bei den damit verbundenen Fallen von Pseudosclerose, zugleich ein Beitrag zur Lehre von der Pseudosclerose (Westphal-Striimpell). Deut. Z . Nervenheilk. 49, 54-73 (1913). R21. RUES,E. M., and Raymunt, J., Influence of estrogens on total serum copper and caeruloplasmin. PTOC.SOC.Exptl. Biol. Med. 92, 465-466 (1956). S1. Sachs, A., Levine, V. E., Hill, F. C., and Hughes, R., Copper and iron in human blood. A.M.A. Arch. Internal Med. 71, 489-501 (1943). S2. Sala, I., and Gambara, L., Contenuto in rame e attivita ossidasica del plasma materno e funicolare. Latlante 28, 458470 (1957). S3. Sandell, E. B., “Colorimetric Determination of Traces of Metals,” 3rd ed. Wiley (Interscience), New York, 1959. 54. Sanders, B. E., Miller, 0. P., and Richard, M. N., Preparation of ceruloplasmin. Arch. Biochem. Biophys. 84, 60-62 (1959). S5. Sands, R. H., and Beinert, H., The function of copper in cytochrome oxidase. Biochem. Biophys. Res. Commun. 1, 175-178 (1959). S6. Sarzeau, A., Sur la prbence du cuivre dans les vbgbtaux et dans le sang. J . Pharm. &%.Accessoires 16, 505 (1830). 87. Sass-Kortsak, A., Jackson, S. H., and Charles, A. F., Studies on ceruloplrtsmin. Vox Sanguinis 6, 87-88 (1960). 58. Sass-Kortsak, A., Baker, R. G., Cameron, H. H., and Neumann, P. Z., The fate of a single intravenous dose of copper in patients with Wilson’s disease. Gastroenkrobgy 42, 229 (1902). 59. Sass-Kortsak, A., Cherniak, M., Geiger, D. W., and Slater, R. J., Observations on ceruloplasmin in Wilson’s disease. J . Clin. Invest. 88, 1672-1682 (1959). S10. b K o r t e a k , A., Glatt, B. S., Cherniak, M., and Cederlund, J., Observations on copper metabolism in homozygotes and heterozygotes of Wilson’s disease. I n “Wilson’s Disease, Some Current Concepts” (J. M. Walshe and J. N. Cumings, eds.), pp. 151-167. Thomas, Springfield, Illinois, 1961. S11. Sayers, R. R., “Copper and Health,” 3rd ed. Copper and Brass hsoc., New York, 1951. S12. Schaffner, F., Sternlieb, I., Barka, T., and Popper, H., Hepatocellular changes in Wilson’s disease. Histochemical and electron microscopic studies. Am. J . Pathol. 41, 315-323 (1962). S13. Scheinberg, I. H., Copper metabolism. Federation Proc. 20, 179-185 (1961). S14. Scheinberg, I. H. Personal communication (1964). 515. Scheinberg, I. H., and Gitlin, D., Deficiency of ceruloplasmin in patients with hepatolenticular degeneration (Wilson’s disease). Science 116, 484-485 (1952). S16. Scheinberg, I. H., and Morell, A. G., Exchange of ceruloplasmin copper with ionic CUMwith reference to Wilson’s disease. J . Clin. Invest. 86, 1193-1201 (1957). 517. Scheinberg, I. H., and Sternlieb, I., Copper metabolism. Pharmacol. Rev. 12, 355380 (1960). 518. Scheinberg, I. H., Aisen, P., and Morell, A. G., Recent studies on ceruloplasmh. Voz Sanguinis 4, 69-70 (1959). S19. Scheinberg, I. H., Cook, C. D., and Murphy, J. A., The concentration of copper
n,
64
ANDREW SASS-KORTSAK
and ceruloplasmin in maternal and infant plasma a t delivery. J . Clin. Invest. 95, 963 (1954). 520. Scheinberg, I. H., Harris, R. S., Morell, A. G., and Dubin, D., Some aspects of the relation of ceruloplasmin to Wilson’s disease. Neurobgy 8, Suppl. 1, 44-51 (1958). 521. Echeinberg, I. H.,Morell, A. G., Harris, R. S., and Berger, A., Concentration of ceruloplasmin in plasma of schizophrenic patients. Science 126, 925-926 (1957). 522. Schulman, S.,and Barbeau, A., Wilson’s disease: A case with almost total loss of cerebral white matter. J . Neuropathol. and Ezptl. Neurol. 22, 105-119 (1963). S23, Schultze, M. O.,Elvehjem, C . A., and Hart, E. B., Further studies on the availability of copper from various sources as a supplement to iron in hemoglobin formation. J . Bwl. Chem. 116, 453457 (1936). S24. Seven, M. J., and Johnson, L. A. (eds.), “Metal-Binding in Medicine,” Group V, pp. 259-296. Lippincott, Philadelphia, Fennsylvania, 1960. S25. Shapiro, J. R.,Morell, A. G., and Scheinberg, I. H., A copper protein of human liver. J . Clin. Invest. 40, 1081 (1961). 526. Shields, G. S.,Marltowitz, H., Klassen, W. H., Cartwright, G . E., and Wintrobe, M. M., Studies on copper metabolism. XXXI. Erythrocyte copper. J . Clin. Invest. 40, 2007-2015 (1961). 927. Siemerliig, E.,and Oloff, H., Pseudosclerose (Westphal-Striimpell)mit Cornealring (Kayser-Fleischer) und doppelseitiger Scheinkatarakt, der nur bei seitlicher Beleuchtung sichtbar ist und der dem nach Verletzung durch Kupfersplitter entstehenden Katarakt ahnlich ist. Klin. Wochschr. 1, 1087-1089 (1922). 528. Silverberg, M., and Gellis, S. S., The liver in juvenile Wilson’s disease. Pediatrics SO, 402-413 (1962). S29. Sjollema, B., Kupfermangel ale Ursache von Krankheiten bei Pflanzen and Tieren. Biochem. 2.267, 151-156 (1933). S30. Sjollema, B., Kupfermangel als Ursache von Tierkrankheiten. Biochem. 2. 296, 372-376 (1938). S31. Smales, A. A.,Mapper, D., and Wood, A. J., The determination by radioactivation, of small quantities of nickel, cobalt and copper in rocks, marine sediments and meteorites. AnaZyst 82, 75-88 (1957). New spe532. Smith, G.F., and McCurdy, W. H., 2,9-Dimethyl-l,lO-phenanthroline. cific in spectrophotometricdetermination of copper. Anal. Chem. 24,371-373 (1952). S33. Smith, G.F., and WilkinS, D. H., A new colorimetric reagent specific for copper. Anal. Chem. 26, 510-511 (1953). S34. Smith, G.F.,McCurdy, W. H., and Diehl, H., Colorimetric determination of iron in raw and treated municipal water supplies by the use of 4,7-diphenyl-1,10phenanthroline. Analyst 77, 418-422 (1952). 535. Soothhill, J. D., Blainey, J. D., Hall, G. S., Neale, F. C., Fischer-Williams, M., and Melnick, 8. C., A family study of the biochemical defects of Wilson’s disease. I n “Wilson’s Disease, Some Current Concepts” (J. M. Walshe and J. N. Cumings, eds.), pp. 124-132. Thomas, Springfield, Illinois, 1961. S36. Spiller, W. G.,The family form of pseudo-sclerosis and other conditions attributed to the lenticular nucleus. 3. Nervous Mental Disease 43, 23-36 (1916). S37. Stark, G. R.,and Dawson, Ch. R., Spectrophotometric microdetermination of copper in copper oxida~esusing oxalyldihydrazide. Anal. Chem. 30, 191-194 (1958). S38. Stein, W.H., Bearn, A. G., and Moore, S., The amino acid content of the blood and urine in Wilson’s disease. J . Clin. Invest. 93, 410-419 (1954).
COPPER METABOLISM
65
539. Steinbuch, M.,and Quentin, M., Preparation of ceruloplasmin. Nature 183, 323324 (1959). S40. Sternlieb, I., and Janowitz, H. D., Absorption of copper in malabsorption syndromes. J . Clin. Invest. 43, 1049-1055 (1964). 541. Sternlieb, I., and Scheinberg, I. H., Penicillamine therapy for hepatolenticular degeneration. J . A m . Med. Assoc. 189, 146-152 (1964). S42. Sternlieb, I., Morell, A. G., and Scheinberg, I. H., The effect of intravenously administered ceruloplasmin on copper absorption in a patient with Wilson’s disease. J . Clin. Invest. 37, 934 (1958). 543. Sternlieb, I., Morell, A. G., and Scheinberg, I. H., Homozygosity and heterozygosity in Wilson’s disease. I n “Wilson’s Disease, Some Current Concepts” (J. M. Walshe and J. N. Cumings, eds.), pp. 133-140. Thomas, Springfield, Illinois, 1961. S44. Sternlieb, I., Morell, A. G., Tucker, W. D., Greene, M. W., and Scheinberg, I. H., The incorporation of copper into ceruloplasmin i n viva: Studies with copper64 and copperm.J . Clin. Invest. 40, 1834-1840 (1961). 545. Sternlieb, I., Morell, A. G., Bauer, C. D., Combes, B., de Bobes-Sternberg, S., and Scheinberg, I. H., Detection of the heterozygous carrier of the Wilson’s disease gene. J . Clin.. Invest. 40, 707-715 (1961). S46. Stoner, R. E., and Dasler, W., Spectrophotometric determination of copper following extraction with 1,5-diphenylcarbohydrazidein benzene. A d . Chem. 32, 1207-1208 (1960). S47. Streifler, M.,and Feldman, S., Effect of dimercaprol (BAL) in hepatolenticular degeneration. A . M . A . Arch. Neurol. Psychiat. 69, 84-90 (1953). 548. Sturgeon, P., and Brubacker, C., Copper deficiency in infants. A syndrome characterized by hypocupremia, iron deficiency anaemia and hypoproteinaemia. A.M.A. J . Diseases Children 92, 254-265 (1956). 549. Sunderman, F. W., Jr., White, J. C., Sunderman, F. W., and Lucyszyn, G. W., Metabolic balance studies in hepatolenticular degeneration treated with diethyldithiocarbamate. Am. J . Med. 34, 875-888 (1963). T1. Thiers, R. E., Contamination in trace element analysis. In Methods Biochem. Anal. 6, 273-335 (1957). T2. Thiers, R. E , and Vallee, B. I,., Distribution of metals in subcellular fractions of rat liver. J . Bid. Chem. 226, 911-920 (1957). T3. Tipton, I. H., and Cook, M. J., Trace elements in human tissue. Part 11. Adult subjects from the United States. Health Phys. 9, 103-145 (1963). T4. Tipton, I. H., Cook, M. J., Steiner, R. L., Boye, C. A., Perry, H. M., Jr., and Schroeder, H. A., Trace elements in human tissue. Part I. Methods. Health Phys. 9, 89-101 (1963). T5. Tompsett, S. L., The copper content of blood. Biochem. J . 28, 1544-1549 (1934). T6. Tompsett, S. L., Factors influencing the absorption of iron and copper from the alimentary tract. Biochem. J . 34, 961-969 (1940). T7. Tousimis, A. J., and Adler, I., Electron-probe X-ray microanalyser study of copper within Descemet’s membrane of Wilson’s disease. J . Histochem. Cytochem. 11, 4047 (1963). T8. Turkington, R.W., and Tracy, F. M., Spectrophotometric determination of ultramicro amounts of copper with 1,5-diphenylcarbohydrazide.A n d . Chem. 30, 16991701 (1958). T9. Turpin, R., Jerome, H., and Schmitt-Jubeau, H., Action de doses progressives de
66
ANDREW SASS-KORTSAK
dibthylstilboestrol sur la cuprbmie chez le rat. Compt. Rend. SOC.BWZ. 146, 17031706 (1952). U1. Ulstrom, R. A., Smith, N. J., and Heimlich, E. M., Transient dysproteinemia in infants, a new syndrome. I. Clinical studies. A.M.A. J. Diseases Children Q2, 219-253 (1956). U2. Underwood, E. J., “Trace Elements in Human and Animal Nutrition,” 2nd ed., pp. 48-99. Academic Press, New York, 1962. U3. Uzman, L. L., Histochemical localization of copper with rubeanic acid. Lab. Invest. 6 , 299-305 (1956). U4. Uzman, L. L., and Denny-Brown, D., Amino-aciduria in hepato-lenticular degeneration (Wilson’s disease). Am. J. Med. Sci. 216, 599-611 (1948). U5. Uzman, L. L., and Hood, B., The familial nature of aminoaciduria in Wilson’s disease (hepatolenticular degeneration). Am. J. Med. Sci. 223, 392-400 (1952). U6. Uzman, L. L., and Jakw, M. A., The Kayser-Fleischer ring. A histochemical and electron microscope study. Neurology 7 , 341-355 (1957). U7. Uzman, L. L., Iber, F. L., Chalmers, T. C., and Knowlton, M., The mechanism of copper deposition in the liver in hepatolenticular degeneration (Wilson’s disease). Am. J. Med. Sci. 231, 511-518 (1956). V1. Valberg, L. S., Holt, J. M., and Szivek, J., Determination of calcium, magnesium, copper and zinc in red blood cells by emission spectrometry. Anal. Chem. 36, 790-792 (1964). V2. Vallee, B. L., Time course of serum copper concentrations of patients with myocardial infarctions. Metab. Clin. ExptZ. 1, 420-434 (1952). V3. Vallee, B. L., Zinc and metalloenzymes. Advan. Protein Chem. 10, 317-384 (1955). V4. Van Ravensteyn, A. H., Metabolism of copper in man. Acta Med. Scand. 118, 163-196 (1944). V5. Vogel, S. F., and Kemper, L., Biochemical reactions of copper within neural mitochondria, with consideration of the role of the metal in the pathogenesis of Wilson’s disease. Lab. Invest. 12, 171-179 (1963). W1. Wainio, W. W., Van der Wende, C. V., and Shimp, N. F., Copper in cytochrome C oxidase. J. BWZ. Chem. 234, 2433-2436 (1959). W2. Walshe, J. M., Current views on the pathogenesis and treatment of Wilson’s disease. A.M.A. Arch. Internal Med. 103, 155-161 (1959). W3. Walshe, J. M., Penicillamine. Practitioner 191,789-795 (1963). W4. Walshe, J. M., Penicillamine, a new oral therapy for Wilson’s disease. Am. J. Med. 21, 487-495 (1956). W5. Walshe, J. M., Studies on the oxidase properties of ceruloplasmin: Factors in Wilson’s disease serum affecting oxidase activity. J. Clin. Incect. 42, 1048-1053 (1963). W6. Walshe, J. M., and Briggs, J. Caeruloplasmin in liver disease, a diagnostic pitfall. Lancet 11, 263-265 (1962). W7. Walshe, J. M., and Carpenter, R. G., Abnormal behavior of Wilson’s disease ceruloplasmin in the presence of enzyme inhibitors. Gastroenterology 48, 499 (1965). W8. Walshe, J. M., and Cumings, J. N. (eds.), “Wilson’s Disease, Some Current Concepts.” Thomas, Springfield, Illinois, 1961. W9. Welshman, S. G., The determination of serum copper. Clin. Chim. Acta 6 , 497498 (1960). W10. Widdowson, E. M., Chemical composition of newly born mammals. Nature 166,62G628 (1950).
COPPER METABOLISM
67
Wll. Widdowson, E. M., McCance, R. A., and Spray, C. M., The chemical composition of the human body. Clin. Sci. 10, 113-125 (1951). W12. Willis, J. B., Analysis of biological materials by atomic absorption spectroscopy. Methods Biochem. Anal. 11, 1-67 (1963). W13. Wilson, J. F., and Lahey, M. E., Failure to induce dietary deficiency of copper in premature infants. Pediatrics 26, 4 W 9 (1960). W14. Wilson, S. A. K., Progressive lenticular degeneration: A familial nervous disease associated with cirrhosis of the liver. Brain 34, 295-509 (1912). W15. Wintrobe, M. M., Cartwright, G. E., and Gubler, C. J., Studies on the function and metabolism of copper. J. Nutr. 60, 395419 (1953). W16. Wohler, F., and Arden, S., Vber den Kupfergehalt menschlicher Organe bei verschiedenen Krankheitssustiinden. KZiu. Wochschr. 41, 509-517 (1963). W17. Wolff, S. M., Renal lesions in Wilson’s disease. Lancet I, 843-845 (1964). Y1. Yamada, H., and Yasunobu, K. T., Monoamine oxidase. 11. Copper, one of the prosthetic groups of plasma monoamine oxidase. J. BWZ. Chem. 237, 3077-3082 (1962). Y2. Yonetani, T., Studies on cytochrome oxidase. 111. Improved preparation and some properties. J. BWZ. Chem. 236, 1680-1688 (1961). 21. Zak, B., and Ressler, N., Serum copper and iron on a single sample. CZin. Chem. 4, 43-48 (1958). 22. Zimdahl, W. T., Hyman, L., and Cook, E. D., Metabolism of copper in hepatolenticular degeneration. Neurology 3, 569-576 (1953). 23. Zipursky, A., Dempsey, H., Markowits, H., Cartwright, G. E., and Wintrobe, M. M., Studies on copper metabolism. XXIV. Hypocupremia in infancy. A.M.A. J. Diseases Children 96, 148-158 (1958). 24. Zwicker, M., Postoperative Serumkupferspiegelveranderungen.Klira. Wochschr. 37, 933-934 (1959).
This Page Intentionally Left Blank
HYPERBARIC OXYGENATION
Sheldon F. Gottlieb Department of Physiology, Department of Anesthesiology, Jefferson Medical College, Philadelphia, Pennsylvania
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
.................. ....................... siological Basis for H ric Oxygenation.. . . . . . . . 2. Some Specific Uses of Hyperbaric Oxygenation.. . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Hypoxia . . . . . . . . . . .................................... 2.2. Respiratory Poisoning.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Myocardial Hypoxia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Asphyxia Neonatorum. . . . . . . . . . . . . . . . . . . . . . . 2.5. Surgery.. . . . . . . . . ...................... 2.6. Shock. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.7. Infectious Diseases. . ......... .......... ....... 2.8. Neoplasia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.9. Hyperoxygenated ................................. 3. Limitations Associated with Pressure and Hyperbaric Oxygenation. 3.1. Physiological Limitations. ................................
Oxygen..................................... .......... Inert Gases.. . . . . . Physical Effects of Pressure.. ...................... ....... Dysbarism. ....... Precautions. ...... ....... ues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. Conclusions. . ..... .................... References....... ..... ................ 3.2. 3.3. 3.4. 3.5. 3.6.
69 69 71 75 75
83 92 94 94 94 109
114 115 119
1. Introduction
1.1. HISTORY
All nature is art, unknown to thee All chance, which thou canst not see; All discord, harmony no€ understood; All partial evil, universal good. Pope During the past 10 years a progressive interest has arisen concerning the use of oxygen under pressure in the treatment of a variety of disease states. The idea of using pressures greater than one atmosphere is not new. Paul Bert’s (B26) classic book on pressure physiology summarized virtually all experience obtained to 1878. In various centers at the end of the eighteenth 69
70
SHELDON F. GOTTLIEB
and beginning of the nineteenth centuries, pressure chambers were constructed for treating a variety of conditions. However the rationale was not easily discernible and the degree of pressure was not of an order of magnitude to result in therapeutic benefit. In 1928 an investigating team (B46) reported the construction of the Timken-Cunningham tank, in Cleveland, Ohio, to be 64 ft in diameter and five stories high. According to this report, Dr. Cunningham claimed that high pressure therapy had been useful in treating a variety of diseases including cancer, syphilis, tuberculosis, pleurisy, diabetes mellitus, and pernicious anemia. Dr. Cunningham never produced records to substantiate his claims, and 100% oxygen was never used. Apparently the tank was dismantled during World War I1 (53). Physiologists since the time of Paul Bert worked intensively on two important aspects of pressure physiology, oxygen toxicity and dysbarism. The dangers of inspired oxygen partial pressures PI,,,)^ greater than that breathed a t one atmosphere were extensively investigated. This work formed the nidus of present day hyperbaric oxygen therapy. The contributions of Haldane (Hl) and de Almeida (D5) deserve special mention. Yet the idea that partial pressures of oxygen greater than one atmosphere absolute might have clinical usefulness does not appear to have germinated until the 1950's. At that time workers in Amsterdam, Glasgow, and London independently and simultaneously conducted experiments that have led to the enormous current interest in this aspect of high pressure physiology. The early work in Amsterdam has been reviewed by Boerema (B32), that of the Glasgow group by Illingworth et al. (Il), and that of the English group by Churchill-Davidson et al. (C14). The resurgent interest in high pressure oxygen therapy in the United States can be traced in part to the early publications of these three European centers and in part to a visit made to this country by Drs. Boerema and Brummelkamp of Amsterdam. Interest in hyperbaric oxygenation increased a t an astounding rate. Even Congress was informed and appealed to for funds (H11). In July 1962 a Workshop on Oxygen Administration under Increased Atmospheric Pressure was held in Washington, D.C., under the auspices of the National Research Council (Al). There have been numerous symposia pertaining to high pressure oxygen therapy on local, regional, national, and international levels. Industries that never before manufactured pressure chambers suddenly became involved in this venture; some have dropped 1 The symbols used throughout the text are consistent with the recommendations of the Pappenheher Committee [Federation Proc. 9, 602-605 (195O)J.Pro, should be read as the inspired partial pressure of oxygen. PBshould be read as the barometric pressure. P , should be read as the partial pressure of oxygen in arterial blood. P b should be read as the partial pressure of oxygen in the alveoli.
HYPERBARIC OXYQENATION
71
it with almost the same suddenness. Hospitals are vying with one another for funds to construct large pressure therapeutic and research centers; having a pressure chamber appears to have become a new status symbol. Is high pressure oxygen therapy a fad that will soon fade? Is there a sound basis for current enthusiasm? What are the prospects for its future? What are the present problems and limitations associated with this new form of therapy? What areas require more intensive investigation if the full therapeutic potentialities of hyperbaric oxygenation are to be realized? It is the purpose of this review to discuss various aspects of this subject in an attempt to answer many of these questions. Before launching into the subject, a word on nomenclature is required. Oxygen under high pressure (OHP) for use in therapy has many synonyms, and numerous abbreviations have been suggested : hyperbaric oxygenation, high pressure oxygen (HPO), high pressure oxygen therapy (HPOT), high oxygen drenching or hyperbaric oxygen drenching (HOD), and hybaroxic therapy. la 1.2. PHYSICAL AND PHYSIOLOGICAL BASIS FOR
HYPERBARIC OXYGENATION A knowledge of the gas laws is an essential prerequisite for comprehending the physical basis of hyperbaric oxygenation. Gases are composed of discrete particles in constant motion moving in straight lines at a high velocity. The pressure or tension of the gas is due to the bombardment of the containing vessel by the constantly moving particles. Pressure may be expressed in a number of ways; some of the more common terms are centimeters of water (cm HzO), millimeters of mercury (mm Hg), pounds per square inch (psi), torr, or atmospheres (atm). Pounds per square inch can be expressed either in absolute pressure or as gauge pressure. Gauge pressures (psig) generally do not take into account the existing ambient atmospheric pressure at sea level. Similar confusion exists for the terminology of atmospheres (atm) ; another designation is atmospheres absolute (ata). It is advisable to refer to pressures in terms of either mm Hg, or pounds per square inch absolute (psia), or atmospheres absolute (ata). The relation between these terms is as follows: la This unnecessary proliferation of terminology also has a counterpart in confusion as to what hyperbaric oxygenation implies. Strictly speaking, any respiratory milieu containing oxygen at a partial pressure greater than 153 mm H g may be co nsidered hyperbaric oxygenation. Generally, however, hyperbaric oxygenation implies use of increased atmospheric pressure, i.e., pressure greater than 760 mm Hg. Thus investigators employing 2 atm 0 2 use the term hyperbaric oxygenation with the same accuracy as those employing 3, 4, or more. However, the physiological, biochemical, pathological, and clinical sequelae will vary markedly, depending on the pressures employed. It would appear that, whenever the term hyperbaric oxygenation or any of its synonyms is used, it sho,uldbe accompanied by a parenthetical value for the specific pressure under consideration.
72 14.7 psia
SHELDON F. GOTTLIEB =
0 psig
=
760 mm Hg = 760 torr
=
1 ata
=
30 inches Hg
One atmosphere is also equal to 33 feet of water. The gas law, based on the work of Boyle, Charles, and Gay-Lussac, states that when temperature is constant, the volume of a perfect gas is inversely proportional to the pressure; when the pressure is constant, the gas volume is directly proportional to the temperature. Avogadro’s Law states that equal volumes of gases a t the same temperature and pressure contain the same number of molecules. The actual number of molecules in a mole of gas is 6.023 X loza, and at 0°C and 760 mm Hg will occupy a volume of 22.4 liters. Dalton’s Law states that each gas in a mixture of gases behaves as if it alone occupied the total volume and exerts a pressure, its partial pressure, independently of the other gases present. Thus the barometric pressure (PB) or total pres. . . P,. The total pressure is equal to the sum sure = PI Pz Pa of the individual partial pressures. Water vapor is a gas and obeys Dalton’s Law. The vapor pressure tendency of a liquid to enter the gaseous phase depends upon the rate of escape of molecules from the surface of the liquid. Gases in contact with water receive water molecules by evaporation until the number of molecules leaving the liquid phase is equal to the number of molecules leaving the gas phase to return to the liquid phase. This evaporation, or escape of molecules from the surface of the liquid, is a function of the temperature of the liquid. ’The temperature of air in the lungs is 37”C, and alveolar air is thought to be in equilibrium with respect to water; the PH,Oat 37°C is equal to 47 mm Hg. Henry’s Law states that the amount of gas dissolved by a given liquid, with which.it does not combine chemically, is directly proportional to the partial pressure of the gas: if the pressure of a gas is doubled then the amount of gas physically dissolved in the solution is doubled. The constant which converts the proportionality to an equality in the Henry’s Law equation is called the Henry’s Law constant: this constant is the solubility coefficient of the gas in the particular solution. The solubility coefficient varies with the nature of the gas and liquid, the presence of solutes in the liquid, and inversely with the temperature. Thus a t a constant pressure, but under hypothermic conditions, more gas can be dissolved in a given amount of fluid (tissue). One of the most important physical laws upon which hyperbaric oxygenation and the decompression tables rests is the law pertaining to diffusion. Diffusion is defined as the passive transfer of a gas through a membrane from a region of high partial pressure to a region of relatively low partial pressure. It is a spontaneous process accompanied by a decrease in free energy; the sole cause of the transfer of the gas from one region to
+ + +
+
HYPERBARIC OXYGENATION
73
another is the inequality of the partial pressure of the gas. The physical factors involved in diffusion were quantified by Graham: the rate of diffusion is directly proportional to the pressure and temperature of the gas and inversely proportional to the square root of the density or molecular weight of the gas. There are many factors that affect the rate of diffusion. The first is the thickness of the membrane through which the gas must diffuse-the greater the distance the slower the rate of diffusion. The second factor is the surface area available for diffusion-the larger the surface area available for diffusion the greater will be the rate of diffusion. The third factor is the solubility of the gas in the medium through which it has to diffuse-the more soluble the gas in the material through which it has to diffuse the faster will be the rate of diffusion. In living tissues this problem is compounded since one must contend with differential solubilities in the various tissues. Edema, by increasing both the distance of diffusion as well as affecting the solubility of the tissues for the gas, imposes a barrier to diffusion. The fourth factor involves the rate at which the gas is brought to the area for diffusion; this is dependent upon the blood supply to the area and would constitute a perfusion limitation. The fifth factor exists primarily for oxygen and is dependent upon the oxygen requirements of the tissues through which the oxygen is diffusing. If the tissues are actively metabolizing and utilizing oxygen, then it can be readily seen that the rate of oxygen diffusion across a given distance will be markedly decreased. At sea level, the normal inspired partial pressure of oxygen is approximately 150 mm Hg. The alveolar partial pressure of oxygen (PA,J is approximately 100 rn Hg. The arterial blood in equilibrium with the alveoli contains very little oxygen in physical solution, approximately 0.3 m1/100 ml plasma. However, the blood contains approximately 20 ml 0 2 / 1 0 0 ml blood. The explanation for the disparity between the volumes of oxygen dissolved in the plasma and blood at the same temperature and pressure is the presence of hemoglobin in the erythrocytes. This pigment has a great affinity for oxygen; 1 g hemoglobin will combine with 1.34 ml oxygen. Thus a normal hemoglobin concentration of approximately 15 g/ 100 ml blood is potentially capable of binding approximately 20 ml oxygen. At a PO,of 100 mm Hg, the hemoglobin is 97-98% saturated with oxygen. When 100% oxygen is administered at 1 ata to a normal individual, the Pao, will be increased and as a consequence so will the PO,of the arterial blood (Pw,). The P,, will rise to approximately 670 mm Hg. The hemoglpbin saturation will be increased by 2-3010 to become fully saturated; the total oxygen content of the blood will increase to approximately 22 ml O&OO ml blood. At the same time the volume of oxygen physically dis-
74
SHELDON F. GOTTLIEB
solved in the plasma will be increased 6-7-fold to approximately 2.0 ml 02/100 plasma (Henry’s Law). If this normal individual continues to breathe 100% oxygen and is now placed in a chamber in which the total ambient pressure is increasing, the PA^, and Pa,, are increased correspondingly. During this compression there is no further increase in the percent saturation of the hemoglobin (since it is already saturated), but the amount of oxygen physically dissolved in the plasma is increased correspondingly (Henry’s Law). At a pressure of 3 ata, for example, the PI^, is 2280 mm Hg, the Pao, is greater than 2100 mm Hg. This corresponds to an approximately 20-fold increase in the P,, as compared to breathing air at 1 ata. In addition, the amount of physically dissolved oxygen is also increased 20-fold to approximately 6.0 ml 02/100 ml plasma. The total oxygen content of the blood will be approximately 26 mlO2/100 ml blood. There is one very important shortcoming of the high oxygen partial pressure associated with physically dissolved oxygen at increased barometric pressures. As the oxygen is removed from solution the PO,decreases a t a very rapid rate. At 3 ata and a normal oxygen consumption, the decrease in PO,from the arterial to the venous end of a capillary would be reflected by a change from 2100 mm Hg 0 2 to 100 mm Hg. In contrast, an equivalent oxygen consumption in the presence of hemoglobin would reflect a change from 100 mm Hg 0 2 a t the arterial end to approximately 60 mm Hg at the venous end of the capillary. The marked difference in changes in the partial pressure of oxygen in the two situations is due to the peculiar properties of the hemoglobin molecule and its ability to bind molecular oxygen. In the presence of a low blood flow and/or abnormally active tissue metabolism, it would be possible to obtain a low venous Po, even though the arterial PO,is very high. If tissues should extract more than 6 volumes oxygen per 100 ml blood, then the 6 volumes of physically dissolved oxygen would not meet the oxygen requirements of the tissue, in terms of either the volume of oxygen available or the partial pressure required for its diffusion. In the case of shunts (failure to oxygenate returning venous blood before its return to the arterial circulation), OHP provides a means of adding sufficient oxygen to that portion of the blood which was normally aerated so as to compensate fully for the venous admixture. However, OHP, i.e., 3 ata oxygen, may provide in some cases just sufEcient oxygen to result in a normal arterial value. In the case of severe shunts, to achieve the arterial oxygen tensions theoretically expected in hyperbaric oxygenation would require using pressures almost twice that now used clinically, i.e., a t least 6 ata would be required. The physical and physiological considerations discussed above as well as
HYPERBARIC OXYGENATION
75
other physiological factors (to be discussed under the appropriate topic) form the foundation upon which clinical hyperbaric oxygenation is built. 2. Some Specific Uses of Hyperbaric Oxygenation
2.1. HYPOXIA Oxygen lack not only stops the machine but wrecks the machinery. Haldane
The implications of the preceding discussion are that, a t the P, greater than 2100 mm Hg that is attained by breathing oxygen at 3 ata, there should be an adequate volume of physically dissolved oxygen (equal to the average arterial-venous difference) as well as a sufficiently high oxygen pressure gradient for oxygen diffusion through the tissues. Theoretically almost any form of hypoxja could be overcome by breathing oxygen at 3 ata. Hypoxia due to carbon monoxide poisoning, hemorrhage, methemoglobinemia, anemia, alveolar diffusion barriers (edema, hyaline membrane disease), right to left cardiopulmonary shunts, ventilation-perfusion inequalities, inadequate tissue perfusion, etc., theoretically should be amenable to hyperbaric oxygen therapy. The use of oxygen under pressure for satisfying the body’s oxygen requirements was demonstrated by the dramatic experiments of Haldane (Hl) 70 years ago and by the more recent experiments of Boerema (B32). Haldane successfully kept mice alive a t 2 ata oxygen while their hemoglobin was completely saturated with carbon monoxide. Boerema has shown that, a t 3 ata oxygen, decreasing the hemoglobin concentration of pigs to less than 1.O% resulted in no electrocardiographic evidence of myocardial anoxia. Certain physiological advantages accrue from eliminating the need for red blood cells; the resulting decreased viscosity could result in an increase in flow and decrease the work of the heart. 2.2. RESPIRATORY POISONING H i s cup of joy i s full when the result of his observations i s put to immediate practical use. Pasteur 2.2.1. Carbon Monoxide
Carbon monoxide is among the leading causes of death from poisonings. Subsequent to the experiments of Haldane, 1 ata oxygen alone or in com-
76
SHELDON F. GOTTLIEB
bination with carbon dioxide was used for the treatment of carbon monoxide poisoning (K8). The use of oxygen under pressure for treating carbon monoxide poisoning seems to have remained dormant until 1942 when End and Long (E3) successfullyused 3 ata oxygen for treating carbon monoxideexposed dogs and guinea pigs. Pitts and Pace (P16) showed that the halflife for the elimination of 20 vol% carbon monoxide in the presence of 2.5 ata oxygen was 19 minutes, in contrast to the half-life of 68 minutes in the presence of 2.5 ata air. Pace et al. (Pl) found that the elimination of carbon monoxide under standard conditions followed the equation COHbt = COHb@t where COHbt is the concentration of CO in the blood at time t , COHbo is the blood carbon monoxide concentration a t time 0, and K is the rate constant. The K values for males and females (30-years old) are, a t 0.2 atm oxygen, 0.0028 (min-’) and 0.0039 (min-I); a t 1.0 ata oxygen, 0.0150 (min-l) and 0.0199 (min-’); at 2.5 ata oxygen, 0.0315 (min-I) and 0.0460 (min-l), respectively. The Glasgow group in recent years has successfully used 2 ata oxygen routinely for treating carbon monoxide poisoning in man (D18, D19, L8, L9, S18,522). The clinical success of this therapy has been confirmed (514). Douglas et al. (D18) have shown that 2 ata oxygen is more efficient than 1ata oxygen plus 5% or 7% carbon dioxide in oxygen for eliminating carbon monoxide from the blood. The least efficient of the gases was 1ata oxygen. The theoretical reason for using hyperbaric oxygenation in carbon monoxide poisoning is based on Le Chatelier’s theorem (if a system a t equilibrium is disturbed, the system will shift in such a way as to minimize the effect of the disturbance) and the fact that it provides a rapid means for overcoming the hypoxemia. High oxygen tensions will favor the association of hemoglobin with oxygen and thereby increase the rate of dissociation of carboxyhemoglobin into its components, reduced hemoglobin and carbon monoxide. By this means the elimination of carbon monoxide is markedly facilitated. The most important aspect of hyperbaric oxygenation in this therapy is the alleviation of hypoxemia. For HPOT to be successful, the patient must be brought to a hyperbaric facility quickly (portable hyperbaric oxygen chambers are available and in use). The experience of the Glasgow group confirms the prediction that once anoxic damage occurs, it will not be reversed by oxygen. The ultimate prognosis of a given carbon monoxide poisoning case will depend upon many factors, not the least of which are the extent and duration of the poisoning prior to institution of therapy. All available evidence indicates that hyperbaric oxygenation is the treatment of choice in carbon monoxide poisoning.
HYPERBARIC OXYGENATION
77
2.2.2. Cyanide There has been some indication in recent years that oxygen may be useful in the treatment of cyanide poisoning (P6).Ivanov (12) showed that 2.8 ata oxygen can restore the cortical electrical activity of mouse cerebrum depressed by KCN. In the United States, Levine (L13) and Cope (C27) suggest the use of 100% oxygen in the treatment of cyanide poisoning. Bond (B33) found that cyanide-induced respiratory inhibition in insects (Sitophilus granarius adults and Tenebroides mauritanicus larvae) “could be partially overcome by increasing the oxygen concentration.” In contrast to mammalian experience, rapid resumption of aerobic metabolism by the insects after cyanide treatment was detrimental to the insect. However, the surprising deleterious effects appeared to be due to one of the products of the aerobic metabolism of cyanide (B34) rather than to oxygen or peroxide formation. The inhibition of catalase by cyanide appeared to have no obvious relationship to the enhancement of toxicity by oxygen. A possible explanation for the beneficial effects of hyperbaric oxygen in the presence of cyanide, permitting survival of tissues and animals, may be the delivery of a sufficient volume of oxygen at a sufficiently high tension to the tissues to provide for a minimal level of extracytochromal metabolism. 2.3. MYOCARDIAL HYPOXIA There are numerous papers citing experimentally induced as well as clinical cases of hypoxia in which hyperbaric oxygen was successfully used to overcome the hypoxia and prevent the occurrence of untoward sequelae -and thereby permit a more rapid recovery. Nelson and Woodward (N2), using Cunningham’s first chamber in Kansas, reported successful use of compressed air (21 psig) in relieving anoxemia caused by either pleurocentesis or injection of gum arabic into the lungs. Hyperbaric oxygen has been used in recent years to overcome the ischemia due to damage to, or occlusion of, peripheral, retinal, cardiac, and mesenteric arteries (11, J1, M9, S9, 517, S19S21, Tl). One aspect of hypoxia which has and still is receiving much attention, because of its potential therapeutic value, is coronary ischemia. Smith and Lawson (S21) reported that the administration of oxygen at 2 ata protected dogs against ventricular fibrillation during acute interruption of the main left circumflex coronary artery. Although the ligation resulted in ischemia to 40% of the ventricular muscle mass, the administration of OHP was able to maintain more of these hypoxic cells in a viable state. In essence these results were confirmed by Meijne et al. (MlO), Petropoulos (P12),
78
SHELDON F. GOWLIEB
Roshe and Allen (R3), and Van Elk and Trippel (V8). Harris and Hitchcock (H4) did not find OHP to effect the release of serum glutamic-oxalacetic transaminase (SGOT; a widely accepted diagnostic aid for clinically assessing suspected cases of myocardial infarction) in dogs following ligation of the anterior descending branch of the left coronary artery. On the surface, these data would tend to be at variance with the data indicating a protective effect on the heart against ventricular fibrillation. It appears, on the basis of Harris and Hitchcock’s work, that destruction of heart muscle occurs following acute myocardial infarction even in the presence of OHP. One does not know whether OHP protected some of the cells in the ischemic area against hypoxic destruction. The question arises of how much myocardial mass must be ischemic (hypoxic) to trigger fibrillation. What degree of ischemia (hypoxia) induces fibrillation? What is the importance of the site of ischemia (hypoxia) in inducing fibrillation? What degree of ischemia will result in destruction of cardiac muscle? What degree of ischemia will permit the release of SGOT into the circulation and still prevent fibrillation? Harris and Hitchcock’s results possibly indicate that less ischemia is required for SGOT release than for triggering fibrillation. Hunter el al. (H16) were unable to confirm the findings of the beneficial effects of oxygen as reported by other investigators, but point out that their techniques for producing myocardial ischemia differed somewhat from those of other investigators. However, they feel, “There is sound theory for the idea of placing a human patient with ischemic myocardial damage under compression.” The importance of collateral circulation for the possible successful use of OHP in cases of myocardial infarction is stressed by Harris and Hitchcock as well as by Meijne et al. (M13). The role of the Thebesian vessels in maintaining myocardial metabolism should be considered. Since the arterial pressure distal to an occlusion approaches 0 mm Hg and the venous pressure, though low in absolute numbers, is higher than the arterial pressure, one can envision a retrograde circulatory oxygen supply to the hypoxic tissues. The eaciency of this venous blood in providing oxygen for ischemic tissues is markedly increased by hyperbaric oxygenation. The promising results obtained in animal experimentation unfortunately have not had their counterpart in controlled clinical studies (Cl): the mortality figures for the control and experimental groups were virtually identical. The marked variance between the theoretical and the practical requires an explanation. The most obvious factor with which to begin the search is the administration of oxygen, and here one finds the technique of administering oxygen to have been inefficient. If one assumes no dif-
HYPERBARIC OXYGENATION
79
ference between the techniques for administering oxygen to the cardiac cases and to the infectious disease cases a t the same institution, one can readily see that the arterial oxygen tensions attained in cardiacs were much lower than the theoretically calculated maximum for 2 ata (Kl). Improvement in the techniques of oxygen administration coupled with higher ambient pressures should result in improved clinical results. Further, the use of hyperbaric oxygenation should be coupled with techniques for improving the rate of opening of collateral circulatory pathways.
NEONATORUM 2.4. ASPHYXIA As has been emphasized, regardless of the underlying cause, oxygen must be supplied to the tissues at a pressure of either one or more atmospheres in order to prevent both the early and later sequelae of acute hypo&. Other areas in which hyperbaric oxygen may have potential use are the treatment of the respiratory distress syndrome of the newborn (RDS) and asphyxia neonatorurn. Hutchison et al. (H17) did not find hyperbaric oxygen per se to be of significant value in the treatment of RDS; correction of the acidosis was superior to hyperbaric oxygenation in increasing survival: 1 of 6 patients treated with OHP survived, whereas of 27 patients in which the biochemical disturbance was corrected 16 survived; of the 16 survivors, 5 had also received oxygen under pressure. The finding of Hutchison et al. (H17) that hyperbaric oxygenation is not of value in treating suspected cases of respiratory distress syndrome of the newborn was recently confirmed by Cochran et al. (C18a). Hyperbaric oxygenation may relieve the hypoxia but has no effect on the respiratory acidosis. In addition it does not effect a more rapid dissolution of the hyaline membrane. A complex disease entity such as the respiratory distress syndrome would seem to require a therapeutic approach that combines some of the individual techniques currently used in various centers for the treatment of this disease. This combined therapy is envisaged as including the use of intermittent positive pressure breathing in conjunction with hyperbaric oxygenation, hypothermia, and proteolytic enzyme therapy. As more is learned about the chemical nature of the pulmonary surfactant, it too can be included in the therapeutic regimen. Although hyperbaric oxygenation may have to be limited because of the ever present dangers of oxygen toxicity (including retrolental fibroplasia), intermittent positive pressure breathing combined with hypothermia and proteolytic enzyme therapy can be continued a t one atmosphere pressure in the absence of the pressure chamber. Although this discussion centered on the mechanical aspects of ventilation, consideration must be given to the role of the
80
SHELDON F. GOTTLIEB
circulation in the pathology of the disease. Any therapeutic advances made in the perfusion aspect of the disease should be considered for incorporation into the combined therapeutic regimen. The use of hyperbaric oxygen for resuscitation of the newborn has created much controversy. Moss (M20) treated 20 infants with OHP. Apparent success was reported in 2 cases of neonatal asphyxia and in 3 other cases treated within 3 days of birth; what happened to the other 15 cases is not mentioned. Hutchison et al. (H18) concluded from their experiences in treating 65 infants, “Apnea neonatorum is not an emergency in which a controlled trial would be permissible, but it is our firm impression that hyperbaric oxygen is the most effective method of resuscitation yet devised for the severely asphyxiated newborn infant.” The work has been criticized by Barrie (B8) and Davis and Tizard (D2) and defended by Hutchison et al. (H19) and Donald (D15). Cross et al. (C31) recently compared the efficacy of hyperbaric oxygenation and intermittent positive pressure breathing (IPPB), using 100% oxygen for resuscitation of asphyxiated mature fetal rabbits delivered by caesarean section. Their data clearly indicate that IPPB was superior to OHP for resuscitative purposes: of 17 rabbits treated with 4 ata oxygen there were no survivors, whereas 10 of 12 animals survived when given IPPB. These experiments also indicated that, contrary to the viewpoint of Goodlin (G14, G15) and Hutchison et al. and in agreement with the comments of Davis and Tizard (D2), cutaneous diffusion of oxygen was not of sufficient magnitude to influence the oxygenation of the animal. Hutchison et al. (H20) questioned the relevance of Cross’s results to asphyxia neonatorum by claiming, “The efficacy of hyperbaric oxygen as a method of resuscitation in the human infant can only be decided by observations made on human infants under everyday conditions of obstetric practice. These conditions have not been sufficiently closely simulated in animal experiments to justify direct extrapolation of the results to humans.” In contradiction to a previous statement (vide supra), Hutchison et al. stated plans to conduct a controlled experiment using human infants. A number of investigators (C30, D3, G25) subsequently pointed out that asphyxia implies not only hypoxia, but also respiratory acidosis, and that IPPB mill not only improve oxygenation, but, in contrast to hyperbaric oxygenation, will assist respiratory exchange. In addition to pointing out that consideration should be given in the treatment of asphyxia neonatorum to the use of hypothermia alone or in conjunction with IPPB, with or without compression to 2, 3, or more ata, Gottlieb (G25) emphasized the importance and relevancy of comparative physiological studies.
HYPERBARIC OXYGENATION
81
2.5. SURGERY
Skepticism i s a useful tool of the inquisitive mind, but it i s scarcely a method of investigation. Thurber The hypoxia resulting from deliberate occlusion of blood vessels during certain surgical procedures severely limits the time surgeons have available for performing connective or palliative operations. Boerema et al. (B30) in 1956 introduced the use of high pressure oxygenation to help ameliorate many of the technical difficulties by prolonging the safe period of circulatory arrest. In order to prolong the margin of safety for circulatory arrest, they suggested increasing the oxygen reservoir of the body by inhalation of oxygen at 3 ata. It would be theoretically possible, by combininghyperbaric oxygenation with hypothermia, to maintain relatively normal tissue oxygen tensions despite a diminished blood supply. Hypothermia not only increases the solubility of the gas, thereby permitting a given mass of tissue to hold a greater volume of gas, but decreases the tissue demand for oxygen; there is an approximately 6% decrease in oxygen coilsumption per degree fall in body temperature. The decreased metabolism also minimizes the chances of metabolic acidosis (providing shivering is prevented). OHP has an additional value in conjunction with the extracorporeal circulation; it increases the oxygenation capacity of the oxygenator and permits smaller priming volumes to be used. Further benefits accrue from combining these techniques: the safety of procedures requiring low blood flow techniques is markedly increased, deep levels of hypothermia may be employed without irritation to the myocardium, and adequate oxygenation should allow for an undisturbed rewarming of the body. Experimentally and clinically, hyperbaric oxygenation has been used in conjunction with hypothermia and extracorporeal circulation for numerous successful corrective and palliative cardiovascular procedures (B4, B25, B29, M14, R2, W4). Meijne (M12) emphasizes the need to separate the value of OHP during the actual period of cardiac arrest and the value of OHP “after the arrest period during lowered cardiac output.” Because of the moderate oxygen stores of the body at 3 ata and the requirement for the large oxygen tension gradients between the capillaries and the mitochondria, he recommends that OHP “be used mainly in conditions where at least some circulation is present.”
82
SHELDON F. GOTTLIEB
2.6. SHOCK
To understand the nature of disease is the fundamental object of medicine, for knowledge about a thing i s the best way to acquire power over it. Selye Hypoxia is one of the important factors that intensify the development of shock. The decreased blood pressure results in the production of a general ischemic hypokinetic anoxia (stagnant anoxia). Secondary anoxic effects such as hypotonic anoxemia only tend to compound the pathological effects of the shock syndrome. Boothby et al. (B36) emphasized the importance of oxygen administration to prevent the deleterious effects of the hypokinetic anoxia. On the basis of this suggestion, Wood et al. (W14) found 100% oxygen at 1 ata to be effective in treating dogs with mild peripheral circulatory failure produced by hemorrhage, trauma, or histamine injection. Probably because it can be more easily controlled, the most consistent results were obtained with hemorrhagic shock. Schnedorf and Orr (S5, S6) also reported on the beneficial action of 1 ata oxygen on the course of hemorrhagic shock produced in dogs. These studies stimulated Frank and Fine (F7) to study the effectiveness of oxygen under pressure in preventing the course of events in hemorrhagic shock. They found that preventing venous anoxemia by the administration of 3 ata oxygen in no way altered the events of the shock and concluded, “Oxygen therapy in shock is therefore of doubtful value.” It was not until 1959 that OHP was resurrected for the treatment of shock. Burnett et al. (B47) found that 2 ata oxygen did not improve the survival of rats subjected to traumatic or histamine shock; it did protect rats during the acute hemolytic crises produced by glycerol. Esmond et al. (E4) and Moulton et al. (M21) found that 3 ata oxygen is beneficial in increasing survival of dogs subjected to experimental hemorrhagic shock and rats following Noble-Collip drum shock. The beneficial effects of OHP on the subsequent events of shock were confirmed in subsequent experiments (A12, C28, C29) and were attributed to the maintenance of the cellular oxidative processes despite the reduction of the normal oxygen transport mechanism. There are conflicting reports concerning the value of OHP in endotoxin or bacteremic shock. Evans et al. (E7) found OHP to be beneficial in improving survival of dogs subjected to Escherichia coli endotoxin shock, whereas Blair et al. (B27) found no beneficial effects of oxygen on bacteremic shock (instillation of saline suspension of feces) in dogs; in other experiments they found that OHP increased survival of Noble-Collip drumshocked rats andlhemorrhage-shocked dogs.
HYPERBARIC OXYGENATION
83
The reasons for the discrepancies in the results between investigators is difficult to ascertain. Perhaps the manner in which the shock was produced, the extent and duration of the shock, the technique of administering the oxygen, and possible differences in premedication of the animals account for the differences. Some of these possibilities derive strong support from the work of Weale (W3), who found that in late stages of shock oxygen therapy becomes ineffective. Also Burnett’s group used 2 ata oxygen whereas Esmond’s group used 3 ata oxygen. In the case of endotoxin shock, Evans et al. employed an endotoxin obtained from a single organism whereas Blair et al. used a complex mixture of materials. Other marked differences were noted in the technique of anesthetization and maintenance of animals : Evans employed pentobarbital anesthesia and artificially ventilated the animals with a respirator via an endotracheal tube, thus assuring adequate oxygenation, whereas Blair et al. employed chloralosed animals and did not assist ventilation. There is need for more detailed studies using standardized techniques (anesthesia, premedication, and assisted ventilation should be used whenever possible) before final judgment is made on whether OHP has beneficial effects on bacteremic shock. The utilization of standardized techniques should simplify comparison of results from different laboratories and of the effects of oxygen in different types of shock, as well as aiding in the identification of underlying differences in the pathophysiology of different types of shock. 2.7. INFECTIOUS DISEASES The area in which hyperbaric oxygenation has produced the most dramatic results is probably in the treatment of anaerobic infections. The theory behind the use of oxygen as an antimicrobial agent is based on the fact that certain species of microorganisms are unable to grow in the presence of nascent oxygen. Therefore oxygen is applied to the site of inflammation to destroy the invading parasite. This was the basis for earlier oxygen and hydrogen peroxide treatments (vide in&) and for the present use of hyperbaric oxygen in the treatment of anaerobic infections. Boerema and Brummelkamp pioneered the development of present-day hyperbaric oxygenation in the treatment of clostridial infections (B31). They reasoned that an effective mode of therapy would be t o envelop the advancing inflammation with oxygen through the capillaries and tissues by virtue of the high pressure oxygen gradient produced by the inhalation of pure oxygen under pressure. The result would be an oxidizing environment deleterious to the invading anaerobic microbe. Cunningham (C33) in 1927 used compressed air for treating syphilis, diabetes, etc. As noted earlier, his work was not well documented and fell into disrepute. However his basic idea for using oxygen in the treatment
84
S H E L D O N F. GOTTLIEB
of syphilis is sound since the causative agent of syphilis, Treponema pallidum, is anaerobic. White (W9) described a case of gas gangrene which he successfully treated with oxygen and hydrogen peroxide. Hoge (H12) described six cases of gas gangrene which he treated with oxygen and hydrogen peroxide; not one case showed extension beyond the oxygen barrier. Hinton (H10) in 1947 described a case of gas gangrene which he successfully treated by the use of oxygen injections. He injected 1000 ml of oxygen every 4-5 hours in order to maintain a barrier around the phlegmon. The use of oxygen under pressure for the treatment of experimental gas gangrene was not encouraging (D6), even though in vitro cultures of Clostridium welchii exposed to 3 atm oxygen were almost completely killed within 3 days. The spores survived this oxygen exposure (PZ). Brummelkamp et al. (B42) reported on the successful treatment of Clostridium perfringens infections in guinea pigs by the use of OHP. In a later series of experiments they were not as fortunate in their results; their data were similar to those obtained by De Almeida (D6). Fortunately, the use of OHP for the treatment of clostridial infections in man was dramatically successful. The clinical work of the Amsterdam group has been summarized (B43, B44). This group administers seven sessions of hyperbaric oxygen therapy in a 3-day period, three exposures to HPOT on the first day and two sessions on each of the next two days. Each treatment consists of exposure to 3 ata oxygen for approximately 2 hours. Neither antibiotics nor anti-gas gangrene serum is used routinely. Antibiotics are used to prevent secondary infections. They prefer to perform any surgery for the repair of necrotic tissue after oxygen therapy, when the patient is nontoxic and improved. The use of OHP facilitated demarcation of the necrotic tissue. The successful use of HPOT in the treatment of clostridial infections has been confirmed by others (M22, 523, W2). The Glasgow group concluded on the basis of their experience that HPOT “is a n aid to, but does not replace, adequate surgery.” They do not mention the treatment of gas gangrene without surgery. Also they used a therapeutic regimen different from that of Boerema’s group; the Glasgow group used 2 ata for 2 hours as compared to the 3 ata used by the Dutch. It would seem that before a group comes to a conclusion about a certain technique or mode of therapy, different from that of those who devised and used it successfully, it should be incumbent upon them to employ the same conditions used by the innovators; otherwise unnecessary confusion results. There is a marked difference between hyperbaric oxygenation at 2 ata as compared with hyperbaric oxygenation at 3 ata. The diffusion of oxygen is greatly enhanced by the extra 760 mm Hg oxygen obtained by increasing
HYPERBARIC OXYGENATION
85
the pressure another atmosphere. This pressure difference may account for the somewhat divergent attitudes of the two groups. In addition, recent data have indicated that the Glasgow patients were not receiving the theoretical amount of oxygen; arterial oxygen tension varied in the range 600-1000 mm Hg (Kl). The need for early and adequate oxygenation can be surmised from the studies of Glover and Mendelson (G12) and Kelley and Page (K6). The use of HPOT in the treatment of tetanus has not met with the same success as for gas gangrene. The rationale for the use of HPOT in the treatment of tetanus is similar to that for gas gangrene. In addition to being caused by an anaerobic organism, the tetanolysin is supposedly oxylabile (W17). Kelley and Page (K6) were unable to show beneficial effects of HPOT in tetanus infection in mice, but were very successful in treating gas gangrene. On the basis of clinical and experimental results to date (B45, L16, P3, P4, 513, W2, W12), the consensus appears to be that HPOT may exert some beneficial effect. Whether any derived benefit is due to the oxygen working alone, or in conjunction with tetanus antitoxin, antibiotics, and sedatives, is at present unknown. In an attempt to determine the mechanism whereby oxygen may be beneficial in the treatment of tetanus, Brummelkamp (B45) oxygenated tetanus toxin by bubbling oxygen through it for 2 hours at 1 and 3 ata. He found no diminution of its toxicity in mice. There is virtually no information at present concerning the oxygen tensions attained in infected tissues when treated with OHP. Since the in vitro lethal PO,for Clostridium welchii is approximately 90 mm Hg, it is surmised that in the clinical cases the minimal tissue PO,is of the same order of magnitude (€345). The deleterious effects of oxygen is not confined solely to obligate anaerobes. Oxygen inhibits the growth of aerobic protista as well (A2, C18, G21, G24, K3, K4, M17, M18, W11). Gottlieb (G20) recently proposed the use of oxygen for the treatment of tuberculosis and Hansen’s disease. This was followed by a detailed report (G24) in which it was shown that in vitro exposure of the tubercle bacillus to 3 ata oxygen for 2 hours twice daily resulted in marked retardation of its growth. The growth of drugresistant strains of Mycobacterium tuberculosis as well as scotochromogenic and Battey-type organisms was also adversely affected. An interesting finding was the enhanced effectiveness of antituberculosis agents in the presence of high oxygen tensions. These effects of oxygen on the tubercle bacillus in in vitro experiments were confirmed in part by the in vivo studies of Allison et al. (A7). Although the in vitro data suggest a direct effect of oxygen on the microbe, the in vivo experimentation must include a consideration of possible physiological and biochemical alterations in the host
86
SHELDON F. GOTTLIEB
which may make the host more or less resistant to invading pathogens and thereby alter the course of an infection. In the case of experimental tuberculosis in rabbits, there was evidence of fewer tubercles in the lungs of oxygen-treated animals. In addition, the tubercles were of smaller size and with less caseation than those of the controls. The possible use of oxygen under pressure for the treatment of sepsis has stimulated further in vitro investigations (H13, M6, N4). The pitfalls encountered in in vitro bacterial experimentation have been discussed in detail by Gottlieb et al. (G23). Some of the earlier findings on oxygen inhibition of bacteria, fungi, and protozoa (A2, C18, K3, K4, M17, M18) should be explored further for possible clinical development. There are a few isolated reports, but no systematic study designed to explore the possible existence of synergistic action between antimicrobial agents and oxygen tension in either in vitro or in vivo systems. In addition to the projected use of OHP for the treatment of Hansen’s disease and tuberculosis, Gottlieb et al. postulated that OHP might be useful in combating other mycobacterial infections as well as infections caused by related organisms, e.g., actinomycetes. Pascale et al. (P4) used OHP for combating a fusiform infection. Nuckolls and Osterhout (N4) showed that OHP will inhibit the growth of Bacteroides species in vitro and in vivo; however, death occurred because of secondary infections. Bacterial infections caused by members of the genus Neisseria might prove amenable to control by hyperbaric oxygenation. Although N . meningitidis, unlike N . gonorrheae, is an aerobic catalase-positive organism, oxygen under increased pressure inhibits other aerobic catalase-positive organisms. In addition, the susceptibility of tissues to oxygen toxicity does not appear to be related to their catalase content (vide infra). If conventional antibiotic therapy is effective in controlling a disease, there is of course no need to employ hyperbaric oxygenation. Hyperbaric oxygenation may be the most propitious therapy for treating the fulminating type of meningitis, often found in military installations. In addition to a direct effect on the organism, hyperbaric oxygen may alter the blood-brain barrier (W16) and make it more permeable to antibiotics. Also, the use of agents that alter cell permeability, e.g. chlorpromazine, in conjunction with OHP and drugs (sensu Zuto) should be explored for facilitating the transport of drugs across the blood-brain barrier. The use of small mammals for studying the effects of oxygen on sepsis presents many difficulties, as demonstrated by the experimentsof Brummelkamp (B45). Small animals are quite prone to the pulmonary damaging effects of oxygen under pressure. Part of this susceptibility may be due to a change in the chemical state of the pulmonary parenchyma due to the
HYPERBARIC OXYGENATION
87
presence of an endemic pneumonia-type condition found in virtually all laboratory strains of small mammals. Another possibility is an exacerbation of the disease process by the oxygen. Such considerations point to the necessity for controlled studies using gnotobiotic animals. The ability of an organism to grow in the presence or absence of nascent oxygen is one of the physiological determinants of bacterial taxonomy. On the basis of the relative susceptibilities of aerobic organisms to grow in the presence of 100% oxygen, a t various total pressures, it is suggested that the concept of “aerobiosis” be extended to include the ability of protista to grow under various oxygen tensions. Such information with its ensuing ramifications may provide another physiological parameter for distinguishing between microbial species (G24). 2.8. NEOPLASIA Another use of hyperbaric oxygenation, which has met with a certain amount of success, is in conjunction with radiotherapy for the treatment of neoplasias. This technique is based on the well-documented observation that normal and malignant tissues are less susceptible to the deleterious effects of X-irradiation when they are relatively hypoxic (G28, G29). This is true not only for mammalian cells, but for lower forms of life as well. The increased sensitivity of cells to X-irradiation in the presence of an increased oxygen tension approaches a factor of 3. This means that a smaller dose of radiation is required in the presence of oxygen to produce an effect equal to that obtained during anoxia. A graphic representation of the oxygen tension-response relationship reveals a very steep rise in the curve with an almost sudden flattening. At the point of flattening there is no further significant increase in tissue sensitivity to X-irradiation in the presence of oxygen. It is at this point also that the increased sensitivity of the tissue to X-irradiation in the presence of oxygen is approximately equal to a factor of 3 (G28). The sensitivity factor is not identical for all cells; Trowel1 (T5) reported a sensitivity of 11 for rat lymph-node lymphocytes. Trowell’s experiments also tend to confirm the well-known observation that sensitivity appears to be better correlated with oxygen availability than with oxygen utilization. According to Gray, the radiosensitivity of cells changes exceedingly rapidly in response to changes in oxygen tension. In some tissues, changes in sensitivity may occur within 2 seconds. The chief function of hyperbaric oxygenation is to provide a sufficiently high oxygen gradient so that by diffusion the entire tumor, especially the relatively hypoxic central portion, will be oxygenated and thereby rendered more susceptible to the damaging effects of X-irradiation. On the basis of the above reasoning, Gray et al. (G29) irradiated Ehrlich
88
SHELDON F. GOTTLIEB
ascites tumor cells in mice breathing oxygen at 3 ata at the time of irradiation. In virtually all cases, tumor regression was greater in animals inspiring oxygen than in animals breathing air. The preliminary studies of Gray et al. stimulated the English group at St. Thomas’ Hospital in London to determine whether the auspicious laboratory experiments can be extrapolated for human use in clinical radiotherapy. Churchill-Davidson, Sanger, and Thomlinson first reported their results on 8 patients in 1955 (C13, Sl). The successful destruction of parts of tumors which received both OHP and X-irradiation as compared to X-irradiation alone inspired them to undertake a more extensive trial using curative radiation doses on whole tumors. Periodic reports of their work have appeared (C14, C15, 52). As of April 1963 they had treated 160 patients with advanced disease; “38 (24%) of them had developed metastases outside the treated area and no less than 10 (6%) had developed a second primary tumor. . . . Although not uniform the overall response to treatment has been better than would have been expected from conventional radiotherapy, and in 15-200/, of the cases has been quite dramatic.” Due to technical difficulties, Churchill-Davidson’s group was unable to use the normal method of irradiation. Emery et al. (E2) overcame these technical problems by developing a portable “Perspex” chamber. Van den Brenk’s (V7) group in Australia has successfully used OHP in conjunction with fractionated megavoltage irradiation in treating advanced malignant disease. Wildemuth (WlO) recently reported his experiences in treating 100 cancer patients and is very favorably disposed toward the combined use of hyperbaric oxygenation and radiotherapy in treating various neoplasias. A note of caution should be introduced. Oxygen administration does not enhance significantly the radiosensitivity of well-oxygenated tissues. Hyperbaric administration of oxygen increases the radiosensitivity of tumors by providing oxygen to an otherwise hypoxic tissue. This implies that hyperbaric oxygenation would increase the radiosensitivity of poorly vascularized normal tissues. Radionecrosis of bone and cartilage is a reality (G29, H14). The dangers of radionecrosis may be overcome by the use of higher energy radiation since comparatively less radiation would be absorbed by bone. Another possible means of circumventing this difficulty would be through the use of a regional perfusion system which can oxygenate the infusion fluid either by an oxygenator or by means of hydrogen peroxide (Ml). Irrespective of the mechanics of oxygenating the infusion fluid, it should be stressed that this type of procedure could be executed at increased ambient pressure. By use of a perfusion technique it may be
HYPERBARIC OXYGENATION
89
possible also to avert many of the toxic side effects of oxygen, alone or in combination with X-irradiation. Many interesting animal experiments essentially confirm the initial animal experiments of Gray et al. and the subsequent clinical experience (G13, G29, T4, V1, V2, V5). Du Sault (D21, D22) recently has shown that the cure rate of spontaneous mammary adenocarcinomas in C3H mice, irradiated while breathing a mixture of 95% oxygen and 5y0 carbon dioxide at 1 ata, is the same as in mice breathing 100% oxygen at 3 ata. He found no increase in normal tissue reactions. In view of these findings he suggests that the technique of choice is irradiation while breathing Oz-COz mixtures at 1ata. The enhanced effectivenessof the 02-COz mixture was postulated as being due to the COz stimulation of respiration and vasodilatation resulting in more oxygen being delivered to the tumor. Unfortunately tissue oxygen tensions were not obtained. In general, the results of radiotherapy in the presence of oxygen have been promising. However, complete “kill” apparently is not obtained, albeit there have been a number of “cures.” Histological and pathological studies suggest that within tumors there are areas which are poorly vascularized and thus poorly oxygenated. The hypoxic environment renders some of the cells relatively radiation-resistant and permits them to survive radiotherapy. The surviving cells are thought to be the cause of later recurrence of tumor growth. Even with the use of hyperbaric oxygenation there may be areas of the tumor that remain relatively hypoxic; in addition, one is dealing with a population of cells with varying degrees of radiosensitivity. The fact that portions of tumors are poorly oxygenated and that, even in the presence of 3 4 ata oxygen, the theoretical oxygen tension in the tumor is not attained has been amply confirmed by oxygen tension measurements made with the oxygen electrode (C9, ClO, C15, E5, E6, V1, V7). At present, there is usually a 15-minute equilibration period of the patient with oxygen prior to irradiation. In the future, pressures greater than 3 or 4 ata may have to be employed to assure adequate tumor oxygenation, and confirmation of the degree of oxygenation by micropolarographic electrodes may be required. Ultimately, such measurements would control the degree of pressurization required for each specific tumor to be irradiated. 2.8.1. Radiomimetics and Other Antitumor Compounds Because of the increased sensitivity of tumors to X-irradiation therapy in the presence of increased tumor oxygen tension, and because nitrogen mustard has an effect on tumors similar to that of ionizing radiation, some investigators have postulated that the tumoricidal effect of cytotoxic drugs
90
SHELDON F. GOTTLIEB
may be enhanced by the presence of an increased oxygen tension in the tumor. Before discussing the subject of radiomimetic drug effects in the presence of high oxygen tensions, one should know first the effect of oxygen alone on the in vivo growth of tumors. De Almeida (D5) found that increasing oxygen tensions results in increasing destruction of the “fuso-cellular sarcoma of Roffo.” In order to increase this differential effect of oxygen between tumors and the host, de Almeida resorted to fasting the animals; fasted animals were found to be less susceptible to oxygen toxicity. Campbell (C3), also using fasted animals, was unable to show oxygen destruction of the Walker rat tumor, mouse sarcoma 63, or the Twort mouse tumor. Kluft and Boerema ( K l l ) , using the mouse malignant mammary tumor TM 8013, found that intermittent hyperbaric oxygenation (2 hours, 3 ata, twice daily) retards the rate of growth of the tumor and depresses lung metastases. However, there was no significant increase in life expectancy of the mice. Antopol et al. (A9) studied the effects of pressure on the growth of carcinoma 755 and sarcoma 180 in mice. They exposed tumor-bearing animals to air at 70 psig for 2-6 hours on 5 consecutive days for 2-5 weeks and found a slower rate of tumor growth. They attributed the slower growth rate to the increased pressure. However, slower tumor growth rate also can be interpreted as being due to the effect of oxygen, rather than to pressure. The puissance of this interpretation is based on the fact that at 70 psig air the Po, is equivalent to that of 1.1 ata 100% oxygen. In light of the possible effects of oxygen on tumors found by others, this interpretation appears to be valid (A10). In order to differentiate between the effects of pressure and the effects of oxygen, it would be necessary to expose one group of tumor-bearing animals to 1.1 ata oxygen (866 mm Hg) and another group to 70 psig of pressure but maintaining the partial pressure of oxygen at 152 mm Hg. It cannot be stated at present what the effects of pressure are on the growth of tumors in vivo or in vitro. Neither can it be said that the increased tensions of the other gases, especially nitrogen, are without effect. Possible synergistic effects of these physical parameters remain to be studied. The conflicting effect of increased oxygen tensions on the destruction of tumors is due in part to the different therapeutic regimens and oxygen tensions employed. The remaining explanations ascribe the different findings to differences in the physiology and biochemistry of these tumors as well as to differences in the unique host-parasite relationship that exists between each tumor and the affected organism. All these factors will have to be studied in sufficient detail before a comprehensive explanation for the successes and failures emerges.
HYPERBARIC OXYGENATION
91
The effect of oxygen in combination with tumoricidal agents has been studied by a number of workers. Gray et al. (G29), in contrast to their results with X-irradiation, did not find the effectiveness of mechlorethamine to be enhanced by the increased oxygen tension. Krementz and Knudson (K12) reported a potentiation by oxygen of the tumoricidal effects of nitrogen mustard (mechlorethamine)on Ehrlich ascites tumor and Sarcoma 37 in mice. Back et al. (Bl-B3), studying the effect of varying oxygen tensions on the in vitro and in vivo action of nitrogen mustard on normal animal tissues and tumor cells, found that the sensitivity to nitrogen mustard in the ascitic and solid forms of sarcoma 180, leukemia 1210, and 6C3 HED Gardner lymphosarcoma remained unaltered even in the presence of 100% oxygen at 4 ata. In addition they were unable to detect oxygen potentiation of cytotoxic effects of nitrogen mustard on the growth rate of Vicia faba bean root tips. The conflicting results reported by Gray (G29) and Krementz and Knudson (K12) and by Back et al. (Bl-B3) have each received support from other workers. Leather and Eckert (L11) found that the maintenance of oxygen saturation of the tumor by OHP a t the same time that nitrogen mustard is infused through the tumor results in an oxygen potentiation of the tumoricidal effects of nitrogen mustard on the spontaneous mammary carcinoma of Swiss-Webster mice. Kinsey (KlO) reported a potentiation of the tumoricidal effects of 5-fluorouracil on the Cloudman S-91 melanoma in DBA/2 mice. In contrast to these positive findings, Marshall et al. (M3) reported no increase in the therapeutic ratio for nitrogen mustard with increases in oxygen tension. However, these experiments were performed a t 1 ata with a partial pressure of oxygen of 0.2-0.6 ata. De Cosse and Rogers (D7), in separate experiments using nitrogen mustard and amethopterin, were unable to detect synergistic effects between these cytotoxic agents and oxygen, 6 hours daily a t 2 ata, on the AMel 4 hamster melanoma. In 6 patients with advanced malignant disease, Adams et al. (A4) found no clinical benefit from combined OHP (2 ata) and chemotherapy. At the present time no concrete generalizations emerge concerning the effectiveness of combined OHP and cytotoxic agents in tumor therapy. Obviously there is room for further exploration. There is no a priori reason for believing that a synergistic action will exist between oxygen and any drug or combination of drugs. Too little is known of the mechanism of action of all these substances and the physiology and biochemistry of the tumors to make valid predictions. Further inquiries into dose-effect relationships between different drugs a t varying oxygen tensions are required. In order to assure concentration of the cytotoxic agent in the tumor, it is suggested that perfusion techniques be used. The use of drugs and radio-
92
SHELDON F. GOTTLIEB
therapy in conjunction with oxygenation may be another fruitful area of investigation. It would be desirable to have oxygen tension (availability) measurements of tumors at the time of therapy. Such information would be invaluable in deciding whether the success or failure of certain forms of therapy could be correlated with tumor oxygen tensions. I n addition, such measurements would permit varying the total pressure from animal to animal to assure a constant tumor oxygen tension. When interpreting data, attention should be given to whether the tumor is a spontaneous tumor or a n implanted tumor; oxygen may exert a n effect on the immunological response of the homograft rejection mechanism. Such possible difficulties may be circumvented by using spontaneous tumors. New insights into immunological mechanisms may be obtained with the implanted tumor technique. Up to this point the discussion has centered on the possible usefulness of oxygen-alone or in combination with radiotherapy or cytotoxic agentsfor the treatment of neoplasias. However, the question arises of the effect, if any, of oxygen alone or in combination with drugs (sensu Zato) on inducing tumors in animals. There is a marked paucity of informat,ion in this area. Heston and Pratt (H8, H9) found that oxygen alone does not increase the rate of spontaneous pulmonary tumors in genetically susceptible strain A mice. However, in the presence of a carcinogen (dibenzanthracene) there is a positive correlation between increased oxygen tension and increased induction of pulmonary tumors. The oxygen had t o be administered at the same time as the drug. Di Paolo (D10, D11) found the incidence of lung tumors to increase in the offspring of pregnant strain A mice given urethan 24 hours prepartum and immediately exposed to a hyperoxic or hypoxic environment.
2.9. HYPEROXYGENATED SOLUTIONS
. . . that eflective research requires an environment that i s not distracted by the immediate needs of industry. Bienfait I n recent years there has been an interest in a new and unique area of respiratory physiology-"breathing fluids." The concepts and experiments to be discussed seem to involve an apparent reversal of evolution-animals breathe a superoxygenated liquid, exchanging oxygen and carbon dioxide as if their lungs were gills (G14, KlkK19, P8). These investigators have shown that mice, rats, or dogs submerged in hyperoxygenated solutions (8-20 atm) will survive for hours, as compared to control animals which die quickly. A balanced salt solution increases the efficiency of this pro-
HYPERBARIC OXYGENATION
93
cedure, as does adding a carbon dioxide buffer such as THAM (tris(hydroxymethyl)aminomethane, 2-amino-2-(hydroxymethyl)-l,3-propanedioI) to the breathing fluid. The implications in drowning are obvious. Indeed, Kylstra recommends that 0.9% saline solution be used in all swimming pools in order to increase the number of successful resuscitations after accidental submersions. Kylstra found that adult white mice, drowned in a balanced salt solution, ceased respiratory movements after 1 minute of submersion at 25°C. If 0.1% hydrogen peroxide is added to the drowning solution the animals will survive for an additional minute or two. On the other hand, animals placed in drowning fluid equilibrated with 8 ata oxygen will survive for 2 hours. On the basis of the observation that animals so immersed make respiratory movements, Kylstra interprets his data to mean that gas exchange in the lungs occurs. Goodlin ascribes survival to diffusion of gases across the skin. A much stronger case is made for pulmonary gas exchange as compared to cutaneous respiration. Pegg et al. (P8) found that rats survived up to 4 hours while spontaneously breathing fluid equilibrated with 5-10 atm oxygen. The expirations were labored, the respiratory frequency was 40% of normal, and the heart rate was 60% of normal. Upon removal from the chamber the animals died within 5 minutes. Letting the animals breathe 100% oxygen in conjunction with intermittent positive pressure breathing (15 cm water) prolonged postsubmersion survival as long as 21 hours. Lungs of animals dying immediately after removal from the chamber were atelectatic and contained serosanguinous fluid. The lung extracts had a surface tension greater than 18 dynes per centimeter, indicating a loss of pulmonary surfactant. Animals that survived for more than 7 hours had hyaline membranes. The loss of pulmonary surfactant can also be inferred from the work of Kylstra and Lanphier (K19), who found that the pressure-volume diagrams of dogs, obtained during inflation of the lung with air following fluid breathing, were suggestive of marked increases in alveolar surface tension. These investigators also reported that fluid ventilation rates of 4200 ml/min and an inspiratory partial pressure of oxygen of approximately 2200 mm Hg resulted in an arterial oxygen content of 14.7-21.6 vol% at rates of oxygen consumption equal to 120 ml/min. Postexposure survival remains a perplexing problem. In the early work, death was probably due to many factors such as exhaustion from the work of breathing, contamination of lungs by urine, feces, and skin oils, atelectasis due to loss of pulmonary surfactant, hyaline membrane, and possibly electrolyte imbalances, especially in the lung parenchyma. However, Kylstra and Lanphier devised techniques to obviate the difficulties en-
94
SHELDON F. GOTTLIEB
countered in complete submersion of the animal in the breathing fluid and thereby eliminated the difficulties from lung contamination by excretory substances. These highly imaginative studies may duplicate some of the changes that occur during the liquid-to-gas transition at birth, and thus may explain some of the complications encountered and possibly provide new insights for the therapy of surfactant deficiency in idiopathic respiratory disease of the newborn. 3. Limitations Associated with Pressure and Hyperbaric Oxygenation
3.1. PHYSIOLOGICAL LIMITATIONS
Science i s built u p with facts, as a house is with stone. But a collection of facts i s no more science than a heap of stones a house. Poincare The pressures that may be required to treat certain forms of hypoxia, i.e., those due to severe shunting, ischemia, or neoplasias, may not be practical. In severe shunts it is conceivable that, in order to alleviate the condition and to hyperoxygenate the patient, pressures greater than the 3 or 4 ata used clinically today would be required. Depending on the high pressure requirement (which may be impractical for a clinical facility) and the duration of exposure (until the abnormal condition is corrected), inconveniently long decompression times might result for the attending personnel in the chamber with the patient. 3.2. OXYGEN Oxygen, although essential for life and given under pressure to help sustain life, under certain circumstances exerts harmful effects. Oxygen toxicity is probably one of the most important limiting factors in high pressure oxygen therapy. Before continuing the discussion, some definitions are in order. Hyperoxia is an increase in the oxygen concentration of the respirable gaseous environment above the usual 150 mm Hg found at sea level. Oxygen toxicity is any variation from the normal structure or function attributable to the action of oxygen which produces deleterious effects. These terms are not interchangeable: the former refers to the physical composition of the gaseous environment; the latter refers to a pathological state of an organism resulting from a change in a specific physical characteristic of the gaseous environment to which the organism was exposed.
HYPERBARIC OXYGENATION
95
Oxygen toxicity is a time-pressure-dependent phenomenon, the two being inversely related (B11). The tolerance to oxygen decreases sharply with increasing pressure. Although the deleterious effects of oxygen at 1 ata (OAP) and OHP share many features, they are not identical. For the sake of brevity they may be considered together. Inspired oxygen concentrations up to 380 mm Hg may be employed for relatively long periods of time without harmful effects (M16). At inspired partial pressures greater than this the physiology of an animal may be disrupted in a number of ways: local irritation of the respiratory tract, derangement of the cerebral nervous system, visual disturbances, circulatory and hemorrhagic disorders, reproductive organ degeneration, complete or partial obliteration of air-containing spaces of the body (lungs, sinuses, middle ear). Local irritation of the respiratory tract is encountered primarily when the PI,,*is greater than 500 mm Hg (526). The studies of Ohlssen (01) have shown that in man breathing 7848% oxygen a t 1 ata, pulmonary symptoms occur between 12-24 hours and include nasal congestion, coughing, substernal stress, sore throat, decreased vital capacity, ear trouble, general fatigue, and mental disturbances. As the PI,, is increased, the time of onset of pulmonary irritability decreases. However, in man oxygen under pressure causes marked cerebral irritability which supervenes and forces the exposure to be terminated before the respiratory aspects become apparent. Small animals (mice, rats, and guinea pigs) die within 4-6 days of continuous exposure to OAP (W5). The initial toxic effects, dyspnea and lethargy, are manifested within 15-20 hours. I n larger animals (dogs) the initial toxic effects occur in 36-42 hours, in the monkey in 72-96 hours. It is interesting to note that animals surviving 240 hours of continuous oxygen exposure show thickening of the pulmonary arterioles. Smith et al. (S16) had found thirty years earlier that rats surviving continuous exposure to 4 ata air (P,$ of 610 mm Hg) showed severe pulmonary arteriosclerotic changes. Interruption of oxygen administration decreases the toxic effects on the pulmonary and cerebral nervous system, especially if the periods of exposure to oxygen are of short duration (K5, P9). However, there is evidence that even routine oxygen therapy may give rise to pulmonary alterations (P18). The predominant involvement of the respiratory tract in oxygen toxicity is probably due t o the fact that the lungs are the first and only organ to come in contact with the full PI,. Dyspnea, capillary congestion, alveolar exudation, hemorrhage, atelectasis, swelling of alveolar walls with encroachment on alveolar spaces, fragmentation of basement membranes between alveolar and endothelial cell layers, accumulation of exudate between the basement
96
SHELDON F. GOTTLIEB
membrane and the alveolar or endothelial cells, immobilization of cilia, appearance of periodic acid-Schiff-positive membranes and changes frequently resembling those seen in human neonatal hyaline membrane disease, and vacuolization of mitochondria constitute the damage done to the lung by exposure to high oxygen tensions (B23, B24, C l l , K2, P18, 58, 524, V3). Present evidence seems to indicate that man may be more resistant to pulmonary damage than are small mammals. Because of the greater sensitivity of small animals to pulmonary damage, studies on the effects of OHP on infectious diseases, neoplasias, and other phenomena are greatly hindered (B45); part of the increased pulmonary susceptibility may be due to the previously mentioned existence of an endemic pneumonia-type condition in these animals In man, as opposed to small animals, pressures greater than 3 ata terminate in convulsions long before pulmonary damage becomes manifest. The precipitation of convulsions constitutes the most striking and inexplicable event of OHP. In small animals, pulmonary damage is noted along with the central nervous system (CNS) involvement. No positive correlations have been shown between the degree of pulmonary involvement and the CNS effects. The CNS symptoms of oxygen toxicity in man and small animals are extremely variable and unpredictable (D16, D17, G22) ; there is as much variation in an individual from day to day as there is between individuals. There is no warning, and under certain conditions there is no detectable lasting damage (D16, D17). The clinical use of oxygen convulsions in the treatment of schizophrenia produced no detectable harmful or beneficial effects (L3). The CNS manifestations of oxygen toxicity may include muscle twitchings of fingers, toes, and mouth, nausea, dizziness, vertigo, nervous irritability (positive Chvostek sign), Jacksonian-type seizures, and grand maltype seizures (D17). Seizure onset may be sudden and not preceded by an aura. Up to the time of seizure, there may be apparent normality, unimpaired judgment, and capacity for work (D16, D17, F9). The cortical electroencephalogram (EEG) during and following a convulsion is indistinguishable from that of grand ma1 epilepsy. In oxygen toxicity, peripheral motor discharges may predominate and cause generalized jactitation without electrical or clinical evidence of cortical disturbance (D16, D17). In Donald’s experiments (D16, D17), those not reaching a convulsive end point showed a tendency toward a slight increase in beta activity and increased synchronization of delta activity in the EEG. Coupled with this was a decrease in the amount and voltage of the dominant alpha frequency (6-12 cps) . Single high-voltage spikes appeared infrequently; these were symmetrical and bilateral.
HYPERBARIC OXYQENATION
97
In small animals, the CNS manifestations of oxygen toxicity have been shown to include foreleg tremors, hindleg tremors, head tremors, or a combination of these as the first visually detectable signs of oxygen toxicity; other animals revealing no apparent overt symptoms of CNS involvement may suddenly manifest mild or violent whole-body seizures (G22, K13). There appears to be no sharp line of demarcation between the effects of oxygen at dif‘ferent inspired concentrations. Although the respiratory and CNS manifestations predominate during relatively short-time exposure to oxygen, longer exposures reveal physiological derangements of other organs and tissues, i.e., visual disturbances, circulatory and hemorrhagic disorders, liver and reproductive organ degeneration, and teratogenic effects (A6, All, B18, C4, D4, D13, F4, G6, G31, M5, M15, R4, 54). The incidence of convulsions in breathing oxygen at 2.8 atm for 30 minutes (U. S. Naval Oxygen Tolerance Test) is approximately 2.00j0.Onset of convulsions has occurred within 15 minutes (F9). Forster and ChurchillDavidson (F6) have shown that subjects at rest will not safely tolerate 4 ata 0 2 for a half hour but are unlikely to show signs of toxicity at 3 ata for the same length of time. Clinical experience with O2 convulsions is difficult to assess. I n Amsterdam, large numbers of patients have been exposed to 3 ata O2 for 2 hours; in Glasgow, patients have been exposed to 2 ata 0 2 for 2 hours. In both centers there have been few ill effects. However, the manner in which the oxygen is administered to the patient is critical. In most cases the type of mask used does not deliver pure oxygen to the patient. An additional d35culty in interpreting clinical results is that patients are often either anesthetized or sedated. These procedures tend to “protect” against the overt CNS symptoms of oxygen toxicity. However, the use of depressants as protective agents (P13) creates a false sense of security. One may only be masking the outward manifestations of oxygen toxicity without preventing cellular damage. That this is likely was shown in the work of Van den Brenk and Jamieson with pentobarbital sodium (V4). These investigators found that pentobarbital sodium accentuated delayed damage to the CNS by OHP while supposedly protecting against convulsions. One might argue that different mechanisms are involved in producing convulsions and permanent damage. Gilbert (G9) has discussed the possibility that a substance may play a dual role, in one instance protect against and in another enhance oxygen toxicity. That oxygen produces permanent motor damage has long been known (B9, B10). Van den B r e d and Jamieson (V6) reported that in over 200 patients treated with 45 psig oxygen in conjunction with radiation for neoplasia and given pentobarbital sodium only 5 cases with convulsions and no deaths were encountered. This incidence of convulsions is similar
98
SHELDON F. GOTTLIEB
to that reported by the U. S. Navy (vide supra). There were no subsequent clinical signs of brain damage, paresis, or psychological disturbances. This is not to imply that permanent CNS damage may not occur. Whether permanent, detectable CNS damage ensues is dependent upon such factors as PoI, duration of exposure, Pco,, circulation, respiration, body temperature, endocrine balance, diet, individual susceptibility, etc. In addition, it is not wise to assume that the body’s reaction to a drug (sensu Zato) would be the same under conditions of hyperbaric oxygenation as under normotensive oxygen conditions (B11). One must approach the use of oxygen under pressure with caution. There are many predisposing factors to oxygen toxicity. Increasing activity or exercise while breathing oxygen under pressure decreases the time required for the symptoms of oxygen poisoning to become manifest (U2, Y2). The reason for the decrease in latency is not known, but may be due to a combination of factors such as increased body temperature, increased sympathetic nervous system activity, and/or increased Pm2. Campbell (C2) has shown that increased ambient temperature decreased the tolerance of rats to oxygen poisoning. These findings were confirmed for rats by Hulpieu and Cole (H15), and for bacteria by Thayson (T2). Popovic et al. (P17) have shown recently that in ground squirrels lowering of body temperature protects against OHP by decreasing metabolism. These data imply that patients with high fever may be more prone to oxygen toxicity than patients with normal temperature; similarly, hypothermic patients should be less susceptible to oxygen toxicity. There is evidence that diet may affect tolerance to oxygen poisoning. Campbell (C2) reported that fasted animals were more likely than nonfasted animals to survive exposure to OHP. This has been confirmed for mice by Gilbert et al. (GS). Endocrine effects also play an important role in oxygen toxicity. As one would expect, oxygen under increased pressure is a stressor. Gerschman and Fenn (G3) found a decreased ascorbic acid concentration in adrenal glands of rats following exposure to 1, 2, 3, and 6 ata oxygen. This was considered evidence that oxygen as a stressor may increase adrenal cortical hormone secretion. These findings were confirmed by Hale et al. (H3). Bean (B12) found that hypophysectomy provided appreciable protection against the acute and chronic adverse effects of OHP; he noted increased latency of onset of neuromuscular reactions, decreased incidence of reactions, decreased severity of reactions, decrease in permanent effects, decreased mortality, and less severe pulmonary damage. Part of the protective action of hypophysectomy could be ascribed to loss of adrenocortical function. Cortisone augments oxygen toxicity (B13). The rest of the protective effect of hypophysectomy is probably attributable to the loss of thyroid
HYPERBARIC OXYGENATION
99
hormone. Grossman and Penrod (G30) found in rats that reduction of thyroid secretion by propylthiouracil afforded protection against 5.5 ata oxygen. Animals fed desiccated thyroid had more severe symptoms. These results were confirmed in rats breathing oxygen at 1 ata by Smith et al. (515). Epinephrine markedly enhances the incidence and severity of the CNS components of oxygen toxicity, as well as intensifying pulmonary damage and increasing mortality (B14). One of the functions of the hypophyseal-adrenocortical axis is generally considered to be a hormonal defense mechanism against a stressor. However in the presence of OHP as a stressor the hypophyseal-adrenocortical axis becomes a liability in that its increased activity augments oxygen toxicity (B12). One of the most important factors affecting an animal’s susceptibility to oxygen toxicity is the partial pressure of carbon dioxide. At low inspired concentrations, carbon dioxide decreases the latency of onset of oxygeninduced convulsions, whereas at higher concentrations it inhibits convulsions (C12). The inhibiting effect of a high PI^% may be related to the anesthetic properties of COZ. The manner in which CO, enhances oxygen toxicity has been extensively studied, but there appears to be no agreement on the mechanism of the effect. Lambertsen et al. (L3) believe the most probable explanation of C02 enhancement of oxygen toxicity to be related to the cerebral vasodilatory properties of COz;dilatation of cerebral blood vessels would result in contact of the tissues with an increased oxygen tension. Gesell (G7) has proposed that the toxic effects of oxygen upon the CNS might be due to autointoxication by COz. The postulated increased tissue COz was supposed to result from the fact that a large supply of physically dissolved oxygen would meet the tissue requirements for oxygen without the usual reduction of hemoglobin. The maintenance of hemoglobin in the oxygenated state would limit the base available for C02 transport, thereby causing COz to be carried at a higher than normal Pco,. The resulting tissue hypercapnia and/or acidosis would be sufficiently severe to cause the symptoms of oxygen toxicity. Gesell did not perform gas analyses on the blood of his experimental animals. However, Lambertsen et al. (Ll, L3) have shown that in unanesthetized man there is no gross accumulation of COzin the brain during oxygen administration even up to the time a t which convulsions begin. Behnke et al. (B19) and Lambertsen et al. (Ll) showed that it was highly improbable that the rise in cerebral COz tension due to the failure of reduction of oxyhemoglobin is s d c i e n t to account for the symptoms of oxygen poisoning. The inhalation of COZin concentrations such that the blood C02 tension is elevated to a degree greater than that observed during OHP does not cause convulsions in man. Lambertsen
100
SHELDON F. GOTTLIEB
et al. (Ll) proposed that the slight endogenous increase in tissue COZ brought about by the absence of reduced hemoglobin formation would serve to protect the animal against oxygen poisoning in contrast to what occurs with exogenously supplied C02.The mechanism of this protection is envisioned as occurring as follows: the increased cellular PCO,would stimulate the respiratory center and cause hyperventilation, which in turn would bring about a decreased arterial PCO,with resultant cerebral vasoconstriction and decrease in cerebral blood flow. The decreased cerebral blood flow would result in a decreased mean brain PO,,thereby protecting the brain against oxygen poisoning. There is some controversy concerning the cerebral vasoconstrictive effects of oxygen. Lambertsen et al. (L2) believe that the vasoconstrictive effect is due to the decreased P a , brought about by hyperventilation (vide supra), rather than by a direct vasoconstrictive effect of oxygen. Jacobson et al. (52) found that, under constant C02 tensions, oxygen exerts a direct cerebral vasoconstrictive effect. The recent work showing that Con-adapted mice were more resistant to oxygen toxicity than unadapted mice (Wl) and the studies on the protective properties of THAM2 (B15, G22, S3) have reopened the question of the role of COz in oxygen toxicity. In addition it should be pointed out that C02 does not appear to enhance the toxic effects of OHP on brain tissue slices (S28) or on the conduction of the nerve impulse in frog sciatic nerve (P10). In most cases, breathing oxygen under pressure, as now used for therapeutic purposes, is not likely to have marked effects on gas transport or acid-base balance. The C02 transport system is unlikely to become a limiting factor in gas exchange for a number of reasons: (1) carbon dioxide is much more soluble than oxygen in body fluids and therefore able to diffuse through tissues more readily; (2) the increased solubility permits C02 to be carried in more significant amounts in physical solution in the plasma; (3) the shape of the COz dissociation curve of oxygenated blood indicates that it is adapted for buffering some changes in Pco,. Theoretically, there is at least one set of circumstances under which maintenance of hemoglobin in the oxygenated state may enhance an acidotic state: in the presence of a severe right-to-left shunt as found in many congenital cardiac anomalies. In the presence of a 50% shunt and OHP, half the venous blood which is acidic and oxygenated will bypass the lungs and go directly into the arterial circulation without having released the metabolically produced C02. Due to the absence of the pH buffering effect of reduced hemoglobin, a profound acidosis may be anticipated.
* THAM = tris buffer [tris(hydroxymethyl)aminomethane].
HYPERBARIC OXYGENATION
101
3.2.1. Mechanism of Oxygen Toxicity There is no result in nature without a cause; understand the cause and you will have no need for the experience. Da Vinci From the time of its discovery, oxygen has been known to exert toxic biological effects (Bll), yet the underlying mechanism remains to be elucidated. Oxygen undoubtedly exerts its toxic effects at either the cellular or subcellular level. One of the more popular concepts is that oxygen reversibly or irreversibly inhibits some key enzyme system(s) in a manner such that serious impairment of essential cellular metabolic functions results. This concept had its origin in the classic studies of Bert (B26) on oxygen inhibition of putrefaction of meat and oxygen consumption of muscle. The early work in this field was reviewed by Stadie et al. (S26). Since the time of Bert, it has been observed that respiration of tissues decreases irreversibly when exposed to OHP (D8, D9, S29). Not all tissues show the same degree of susceptibility to oxygen toxicity. Stadie (529) found in the rat the descending order of susceptibility to be brain > kidney > liver > lung > muscle. Dickens (D8) found the order to be brain cortex (finely divided) > brain cortex (slices) > spinal cord > liver > testis > kidney > lung > muscle. Both groups of investigators agree that brain is the most susceptible tissue while muscle is the most resistant, and that lung tissue is more resistant than other tissues except muscle. This is a surprising finding since in vivo the lungs show marked oxygen effects (wide supra). The finding lends credence to the concept that the respiratory tract is predominantly involved probably because it is the only tissue exposed to the full inspired oxygen tension. The reason for the discrepancy in the relative susceptibility of liver and kidney is not known. In the 1940’sStadies’ group in the United States and Dickens in England undertook a detailed analysis of numerous enzyme systems, and attempted to correlate the speed of onset of the acute CNS symptoms of oxygen toxicity with the rate of inactivation of these enzymes exposed to the same gaseous conditions as the animals. They were unable to correlate the rate of enzyme inactivation with speed of onset of oxygen toxicity; in each system studied, the time required to inactivate the enzymes was greater than the time required for the onset of oxygen toxicity in the intact animal (D8, D9, 527, 5304332). Jamieson and Van den Brenk (54) recently reported a decrease in total dehydrogenase activity in the lung tissue of rats exposed to OHP before overt manifestations of oxygen toxicity appeared. This is the first positive correlation reported between enzyme inactivation
102
SHELDON F. GO!PTLIEB
and time of onset of oxygen toxicity. Because total dehydrogenase activity was measured they were unable to specify a particular enzyme system. However, this is an isolated observation that must be confirmed. The available evidence suggests the possible involvement of a few particularly sensitive enzyme reactions. That general depression of enzyme activity is not involved can be inferred from the fact that many enzyme systems are unaffected by high oxygen tensions, and that not all tissues show equal susceptibility to oxygen toxicity. It is interesting that those enzymes which are affected by oxygen require in general a sulfhydryl group for their functioning. The sulfhydryl group may be either an integral part of the enzyme molecule or a component of a coenzyme (D8, D9, H6, H7). Among the enzymes adversely affected by oxygen and requiring sulfhydryl groups are those involved in oxidative metabolism, i.e., the enzymes of the tricarboxylic acid cycle. In addition, glyceraldehyde-3-phosphodehydrogenase, pyruvic oxidase, and choline acetylase, which require as coenzymes the sulfhydryl-containing compounds glutathione, lipoic acid, and coenzyme A, respectively, have been implicated as possible sites of oxygen inhibition. Pyruvic oxidase is of particular interest and importance because of its central role in carbohydrate metabolism (T3). Gordon et al. (G16) found increased fructose-l,6-diphosphatein liver and muscle of rats but not in the brain following a 20-minute exposure to 6 ata oxygen. Aldolase and glyceraldehyde phosphodehydrogenase measurements in liver and muscle indicated that these enzymes did not become rate-limiting during exposure. The existence of a Pasteur effect was proposed to explain these data. Wood and Watson (W15) found that the concentration of y-aminobutyric acid (GABA) in the brain of rats exposed to 6 ata O2 for 33 minutes was markedly decreased as compared to unexposed controls. The degree of reduction of GABA was related to the severity of symptoms of oxygen poisoning. The levels of glutamic acid, aspartic acid, and total a-amino acids were not altered by exposure to OHP. These investigators proposed that the GABA metabolic shunt would become a major pathway in the tricarboxylic acid cycle of the brain if oxygen inhibits the a-ketoglutaric dehydrogenase step (T3). If this occurs, a greater requirement for GABA would result. If the rate of formation of GABA cannot keep pace with the increased demand, the concentration of GABA in the brain will decrease and impair the functioning of the oxidative metabolic pathway via the GABA shunt. These investigators view the cause of oxygen toxicity as an oxygen inhibition of a-ketoglutaric dehydrogenase and a decreased GABA concentration. This hypothesis is strengthened by the observation
HYPERBARIC OXYGENATION
103
that intraperitoneally administered GABA protects rats against the toxic effects of OHP (W16). Oxygen-induced inhibition of enzymes of carbohydrate metabolism as the primary cause of mammalian oxygen toxicity is strongly supported by comparative biochemical considerations. There is a marked parallelism between the mammalian oxygen-sensitive enzyme systems implicated in the mechanism of oxygen toxicity and the botanical enzyme systems similarly sensitive to oxygen (B5-B7, P19, T6). An alternative proposal to the oxygen-induced enzyme inhibition theory for the mechanism of oxygen toxicity (vide supra) seems to present itself. In view of the fact that oxygen enhances the activity of some enzymes and can function as an enzyme inducer (D12), the enzyme theory should not be centered only around oxygen-caused enzyme inhibition; it also should be concerned with possible derangement of the kinetics of interrelated enzyme systems, caused either by oxygen enhancement of enzyme reactions or by oxygen induction of new enzyme systems or stimulation of further enzyme production of existing and functioning enzymes. The theory concerning the oxidation of essential metabolites, resulting in a relative depletion of key substance(s) to certain critical concentrations, might possibly explain the existence of a preconvulsive latent period in animals subjected to OHP. This hypothesis also could explain individual as well as species differences in susceptibility to oxygen toxicity. Recovery from the convulsion might be explained on the basis of regeneration or resynthesis of the key metabolite(s). This follows from the existence of mechanisms for regeneration of sulfhydryl groups. The hypothesis is strengthened by the fact that the administration of reducing substances (G4,54) or end products of metabolic reactions (W16) serves to prolong the preconvulsive latency in animals exposed to OHP. However, there is a major drawback to these theories: the supporting data were obtained either from in witro experiments or from animals following exposure to OHP. The question arises of whether the decrease either in enzyme activity or in concentration of one or more metabolites is a primary effect of oxygen and thus a possible cause of oxygen toxicity, or a secondary effect due to some other primary disturbance. For example, the possibility that these changes are a result of the convulsion rather than a cause of the convulsion has never been obviated. In addition, it seems to be mandatory that alternative means be found to inactivate the suspected enzyme system@) in order to show that the symptoms and pathology of oxygen toxicity will still become manifest. Any proposed theory of oxygen toxicity must also account for the recovery from the acute convulsive phase of oxygen toxicity
104
SHELDON F. GOTTLIEB
and the variable time of recurrence. Such a theory will have to include information on local blood flows, respiration, gas concentrations, pH, etc. The extrapolation of in vitro data to in vivo occurrences is extremely difficult in regard to the enzymatic mechanisms of oxygen toxicity. One difficulty in relating in vitro data to in vivo occurrences pertains to the PO, of the affected tissues. In vitro preparations exposed to OHP receive the full impact of the high oxygen tension, which may not be mirrored in vivo. Conceivably the affected system may not be receiving the full oxygen tension as reflected in the P,,,. Due to local blood flow, diffusion, tissue consumption, etc., it is possible that the particular system is affected by oxygen tensions hundreds to thousands of Hg less than in the arterial circulation. That the oxygen tension in the brain varies during OHP exposure was shown by Bean (B16). Because an in vitro enzyme system is inhibited by 4, 5, or 6 ata oxygen does not necessarily imply that this is the same enzyme system affected in vivo. Another possibility exists, namely, that just a few cells in a minute localized area of the cortex are adversely affected and that this results in a disruption of the electrical activity of the cortex. This disruption manifests itself as a local or general convulsion. Homogenates would not reveal this because of the complete destruction of the structure-function relationship ordinarily found in the intact CNS. The few cells in the susceptible focal area may be subjected to oxygen tensions varying from a few millimeters of Hg to a few thousand millimeters of Hg. Little is known about the actual oxygen tension required to upset normal cortical function. Because of the presence of catalase in virtually all tissues, it is unlikely that hydrogen peroxide is involved in the mechanism of oxygen toxicity. In addition, Dickens (D8) showed that the susceptibility of tissues to oxygen poisoning is not related to catalase content. Neither was he able to detect peroxide formation in brain tissue exposed to OHP in the presence of cyanide or aside. Cyanide or aside inhibits the breakdown'of hydrogen peroxide by catalase. These observations were essentially confirmed by Bond (B34) for insects. Gilbert and Lowenberg (GlO) found that frog skeletal muscle exposed to 140 atm O2for 2 hours produced the equivalent of 1.03 m2M H202/kg muscle. Unlike the 140 ata 02,the presence of 1 mM HzOz did not seem to influence the resting muscle membrane potential. Wolman and Zaidel (W13) suggested that unsaturated phospholipids and other unsaturated lipids are the primary target of oxygen. Becker and Galvin (B17) found a significant increase in cerebral lipid peroxides (CLP) in rats exposed to oxygen pressure greater than 2 ata. However, the CLP levels did not differ between convulsing and nonconvulsing animals, and at 2 ata, when occasional convulsions occurred, there was no corresponding
HYPERBARIC OXYGENATION
105
increase in CLP. They concluded that CLP elevation probably does not play a major role in oxygen-induced convulsions, but did not dismiss the possibility of local increases in CLP, which would not be detected in whole brain determinations but would be capable of triggering a seizure. In vitamin E-deficient rats, abnormal lipid peroxidation appears to be responsible for the hyperoxic hemolysis (M15). The effect of oxygen on transport mechanisms has received limited attention. Mullins (M23) reported that exposure of hen erythrocytes to 70 atm oxygen decreased potassium ion exchange, but that this oxygen tension was without effect on rabbit erythrocytes. In view of the findings that nitrogen behaved similarly to oxygen, it was concluded that the pressure per se, as opposed to a specific gas effect, was responsible for altering the cell potassium ion exchange. The difference in responsiveness to pressure between the two types of erythrocyte may have been due to differences in metabolism;the hen erythrocyte is metabolically more active than the rabbit erythrocyte. Bruce and Stannard (B40) found that X-irradiation of yeast cells resulted in marked disturbance of the ability to retain potassium ions; this effect was dependent on the presence of oxygen during irradiation. Perhaps such changes in membrane electrolyte permeability may in part account for the lethal effects of radiation in the presence of oxygen in tumor therapy. Direct effects of oxygen on sodium ion transport were observed by Falsetti (Fl), who found that oxygen tensions greater than 8 ata markedly inhibited sodium ion transport across the frog skin. That this inhibition was due to oxygen and not to pressure was shown by the absence of an effect when the experiment was performed under increased nitrogen pressures with the PO,remaining constant at the normal value. These results were interpreted as being consistent with the concept that high oxygen tensions interfere with the oxidative enzyme systems, thereby decreasing the supply of energy available for ion transport. The possibility of a direct interference with the transport mechanism has not been disproven. Stuart et al. (S34) found that exposure of baker’s yeast to high oxygen tensions results in inability to retain potassium ions. Recently, Glinka and Reinhold (G11) have shown that oxygen and carbon dioxide have opposite effects on the movement of water into and out of plant cells. Increased oxygen tensions enhanced cell permeability to water, whereas carbon dioxide acted to decrease water flux. Carbon dioxide partially reversed the oxygen effect of increasing membrane permeability to water. Although none of these investigators studied the effects of gas tensions on membrane permeability to other substances (amino acids, glucose, etc.), there is no evidence to suggest a selective effect on sodium, potassium, and water permeabiljty. In view of the transudations found in the alveoli (vide supra), increased
106
SHELDON F. CtOTTLIEB
oxygen tensions seem to exert marked permeability effects on a wide variety of substances. Indeed, Wood and Watson (W16) found that OHP increases the amino acid permeability of the blood-brain barrier. Gerschman et al. (G5) proposed a provocative theory concerning a common mechanism for radiation injury and oxygen toxicity. The theory is based on the formation of oxidizing free radicals. The role of trace metals in the genesis of oxygen toxicity has been reviewed by Haugaard (H7a). The teratogenic effects of oxygen, alluded to previously, may be due to oxygen breakage of chromosomes (C21a). Conger and Fairchild have shown that chromosomal aberrations in Tradescantia paludosa increase with increasing partial pressures of oxygen. Although the mechanism by which oxygen causes breakage of chromosomes is not understood, the work of Gilbert et al. (G7a) may provide a possible explanation. These investigators found that high oxygen tensions increased the viscosity of deoxyribonucleic acid in the presence of reducing substances, such as reduced glutathione. The underlying mechanism of the local respiratory irritation caused by oxygen, although unknown, is undoubtedly related to an abnormally high PO,.Nonspecific effects from the tanks, pipes, regulators, and masks have been excluded by a study of controls given room air through the same system used to demonstrate pulmonary irritation due to oxygen. No untoward symptoms resulted from this procedure (C21). Questions recently have been raised concerning the role of pulmonary surfactant in the genesis of pulmonary manifestations of oxygen toxicity (C17). Bondurant and Smith (B35) were not able to demonstrate an effect of oxygen on the pulmonary surfactant extracted from rats following exposure of the animals to 6 ata oxygen. However, Collier's (C20) findings, using rabbit lung extracts, support the hypothesis that lung damage in oxygen toxicity is due to a loss of surfactant, which results from damage to the normal mechanism for its production. The fact that a suitable quantitative technique for measuring pulmonary surfactant is as yet unavailable makes results from experimentation on the effects of oxygen on pulmonary surfactant difficult to interpret. In most experiments on OHP and in the clinical use of OHP, the animals and patients breathe 100% oxygen for varying periods of time. Since man has evolved in an atmosphere of high gaseous nitrogen content, during the course of evolution he may have become dependent upon nitrogen and thus in some way this gas may exert subtle physiological effects. On the basis of these considerations, it may be conjectured that the removal of nitrogen leads to deleterious physiological reactions: the inference is t h a t oxygen toxicity may be due to the denitrogenation associated with the inhalation of pure oxygen. However this hypothesis is untenable. Bert (B26) showed
HYPERBARIC OXYGENATION
107
that oxygen toxicity is a function of the partial pressure of oxygen even in the presence of other gases. Dickens (DS, D9) found no protective effect of nitrogen other than as an oxygen diluent in tissue studies. However, it should be pointed out that, although human experiments using 100% oxygen a t total ambient pressures less than 760 mm Hg for relatively short periods revealed no derangement attributable to nitrogen lack (C21), and although Rahn and MacHattie (Rl) demonstrated that mice can live and reproduce in a pure oxygen atmosphere a t reduced pressures, there have been reports that chick embryos will not develop normally if nitrogen is absent or replaced by another chemically inert gas (A5, A6, V l l ) . The observation on the chick embryos has been challenged by Boriskin et al. (B37). 3.3. INERTGASES Experiment i s the sole source of truth. I t alone can teach us anything new, it alone can give u s certainty. Poincar6 Although the absence of an inert gas does not appear to be involved in the mechanism of oxygen toxicity, it should be pointed out that, in the absence of an inert gas in the respirable milieu, atelectatic changes may be favored. However, in clinical hyperbaric oxygenation it is not the absence of nitrogen that is of immediate importance, but rather the presence of nitrogen in concentrations 3 4 times that to which man is ordinarily exposed. This is not so much a hazard to the patient as to the attending personnel. At 3 ata, nitrogen is not exerting much effect (M11). However, pressures greater than 3 ata conceivably will have to be employed, i.e., in treatment of severe shunts or neoplasias (vide supra). Nitrogen at increased pressure has peculiar physiological effects. Cousteau refers to it as “the rapture of the deep.” Lanphier refers to it as Martini’s law-“The mental effects of each 50 feet of descent breathing air are approximately equivalent to those of one dry martini, assuming the stomach to be empty.” The physiological effects of nitrogen as well as its physical properties have been studied extensively (B20-B22). That COZis the cause of nitrogen narcosis was shown to be highly unlikely by the work of Lanphier and Morin (L5) and Lanphier and Busby (L6). The insidious nature of nitrogen narcosis with its resultant impairment of judgment and h e movement may portend grave dangers for surgeons and attending personnel working under certain hyperbaric conditions. Thus the question of appropriate breathing mixtures for medical attending
108
SHELDON F. GOTTLIEB
personnel becomes an important consideration for those contemplating the use of hyperbaric oxygenation. A brief excursion into the physiological effects of the inert atmospheric gases that may substitute for nitrogen as an oxygen diluent appears to be in order. Among these are helium, neon, argon, krypton, and xenon. Hydrogen is generally not included, since in the presence of high oxygen tensions the breathing mixture would be potentially explosive. Conceivably hydrogen may be combined with one or more inert gases and thereby obviate the explosion hazard. Radon, because of its radioactive properties, is an extremely toxic gas and therefore excluded from further consideration. Helium, neon, argon, krypton, and xenon, apparently nonessential to life, produce certain readily recognizable biological effects when the normal inspired partial pressure of the gas is greatly exceeded. Some of the stable isotopes of the helium group of gases act upon organisms, producing narcosis, decreased response to stimuli, metabolic alterations, decreased oxygen-dependent sensitivity to radiation, and altered rate of development. In general the activity of the gases varies in direct relation to molecular weight. Lazerev et al. (LlO), Lawrence et al. (L7), and Cullen and Gross (C32) have postulated and demonstrated the anesthetic properties of xenon in man. Further studies on xenon narcosis were carried out in monkeys (P14), dogs (D14), and rats (G19). The anesthetic potency of xenon is greater than that of nitrous oxide and nitrogen and thus could not substitute for nitrogen in hyperbaric oxygenation. Argon and krypton also exert greater anesthetic properties than nitrogen and thus cannot function as nitrogen substitutes in breathing mixtures for man in high pressure environments. Investigators have observed a number of biological effects of the helium group of elements on the organizational units of the organism, organ, and cell. These effects included altering the rate of metamorphosis in insects (C23, F8), inhibition of muscular contraction (G26), altering the rate of metabolism of the whole animal and tissue slices (C25, C26, L12, L14, S35, S36), protecting cells against the oxygen-dependent effects of X-irradiation (El), protecting animals against electroshock convulsions (C5, C6), and their use as nitrogen replacements in compression of divers and caisson workers (B21). The usefulness of the inert gases as models for studying the mechanism of anesthesia has recently been reviewed by Featherstone and Muehlbaecher (F2). Helium, according to present knowledge, would be the most likely gas to replace nitrogen in a breathing mixture for use in high pressure environments. However, this leads to difficult problems in vocal communication (S10). There is insufficientinformation concerning the physiology of neon to suggest it as a definite alternative to helium and thus dispense with the
HYPERBARIC OXYGENATION
109
communication problem. On the other hand, it is difficult to invoke a strong argument against the use of neon in a breathing mixture. There is no reason to suspect that neon will have peculiar physiological properties out of line with the effects theoretically predictable on the basis of comparative physiology and biochemistry. The advantages of the inert gases, helium and neon, as nitrogen substitutes are their comparatively lower narcotic properties and lower solubilities in body fluids. Their major disadvantage is the tendency toward faster diffusion into bubbles, thereby tending to prolong decompression time. Nitrogen has the primary advantage of requiring shorter decompression time at the pressures currently used in hyperbaric medicine. By using a mixture of one or more of these gases, with the oxygen kept constant at PO,of 150 mm Hg, a product conceivably could be obtained that would provide a respirable atmosphere with the advantages of rapid decompression time and low narcotic ability. Of course the ultimate proportions of the gases used will depend upon the projected time-pressure requirement of the therapeutic procedure. The physiological and biochemical effects of the atmospheric inert gases should not be excluded from consideration in high pressure physiology studies. 3.4. PHYS~CAL EFFECTS OF PRESSURE
What effects can be expected from the increase in pressure per se on human physiology and biochemistry? Increasing pressure should accelerate chemical processes leading to a decrease in volume, whereas any biochemical reaction involving an increase in volume will tend to be inhibited (J6). Reactions involving small molecules most likely will have small volume changes, whereas reactions involving large molecules, such as proteins and nucleic acids, may reveal large volume changes (55). Pressure may affect cell, tissue, or organ function by increasing solation of the cytoplasm (M4). The effects of the increase in cytoplasmic solution may result in acceleration of interaction of particles, denaturation of critical enzymes, or changes in pH (M19). Factors such as these must not be overlooked in the mechanism of oxygen toxicity and inert gas effects (e.g., nitrogen narcosis). 3.5. DYSBARISM Compression and decompression of mammals bring about physical changes that could have serious medical consequences and result in complex legal problems. Compression difficulties may involve both patient and attending personnel, decompression difficulties are primarily a problem
110
GHELDON F. GOTTLIEB
for the attending personnel. Patients will be breathing pure oxygen; dysbarism caused by oxygen is rare due to the rapid metabolic utilization of the oxygen by the tissues. There is no metabolic utilization of inert gases. Dysbarism, a syndrome common to those working in either a pressure or altitude chamber, is a general term connoting adverse effects resulting from changes in barometric pressure. The term includes barotitis, barosinusitis, gastrointestinal distention, air embolism, and decompression sickness (“bends,” caisson disease). The first three abnormalities may be considered to be caused by “trapped” gases, and the remaining symptoms to result from the evolution and expansion of dissolved gases. Any body cavity (paranasal sinuses, middle ear, periodontal abscesses, and intestines) unable to equilibrate with the ambient pressure during either compression or decompression could be a source of serious difficulty. During compression, the greatest volume change occurs within the first 7.5 psig. If the personnel can be compressed to this point without experiencing difficulty, the likelihood is that they can be completely compressed without experiencing untoward reactions. About 5% of the general population cannot successfully equalize pressure in the middle ear or paranasal sinuses. Barotitis can be, and in fact has been, circumvented by myringotomy (F6).Barotitis also can be overcome by swallowing, yawning, or Valsalva maneuvers. Barosinusitis may be severe and if compression is continued result in rupture of the blood vessels with subsequent filling of the sinuses. The significance of these potential complications when choosing staff to man the chamber can be surmised from Boerema’s experiences. He found that one half his staff could not work in the chamber (B32). Problems may be encountered in unconscious patients as well as in infants unable to communicate the reasons for, or sites of, discomfort and pain. During decompression, the expansion of the gases in the closed body cavities may also give rise to severe problems involving the paranasal sinuses, middle ear, improperly inserted tooth fillings, and intestines. Conditions causing inflammation or edema of the mucous membranes may result in obstruction of sinus orifices and eustachian tubes, with consequent failure to equilibrate pressure and development of barotitis and/or barosinusitis. Special attention may have to be given to the expansion of gases in the gastrointestinal tract. According to Boyle’s Law, one would expect the gases to expand during decompression and to enlarge the abdominal circumference. This enlargement will occur until flatus or gaseous eructation causes deflation. Severe abdominal pain will result if normal adjustments do not occur. Either the amount of gas in the gastrointestinal tract is reduced or the individual must be recompressed to an ambient pressure providing relief. This problem may be of particular significance in individuals who have
HYPERBARIC OXYGENATION
111
swallowed significant amounts of air during performance of Valsalva maneuvers on compression. Air embolism is not synonymous with nor a manifestation of decompression sickness. It refers specifically to air embolization of cerebral vessels subsequent to the rupture of lung parenchyma by expanding gases. Air embolism should not be confused with the bubble formation that occurs in the circulatory system and tissues in decompression sickness (vide infra). Air embolism could be expected to result from transient blockage of expiratory pathways by general (closed glottis) or localized gas trapping in the lungs of patients and attending personnel during decompression. The therapy for this type of accident consists of immediate compression to 6 ata. Emphysema patients, asthmatics, persons with lung cysts or tuberculous lesions, etc., may be particularly prone to this phenomenon. That does not imply that such persons may not benefit from hyperbaric oxygenation, if they manifest symptoms of disease states which may be alleviated by OHP; it is implied that extreme caution must be taken during decompression of individuals so afflicted. The statement also implies that individuals exhibiting one or more of these symptoms should be considered a poor risk for the job of attending personnel. Yanda et al. ( Y l ) recently showed that there is no evidence of increased hazard in exposing emphysema patients to pressure if one uses special decompression schedules combined with drug therapy. However, decompression times are relatively long as compared to the standard U. S. Navy decompression tables. They reported a transient (2-28 days) but significant improvement in residual volume and clinical condition of three emphysema patients, and improvement in residual volume of four emphysema patients. Decompression sickness is thought to be the result of intravascular, intracellular, and/or extracellular bubble formation. Bubbles tend to form in any tissue saturated with an inert gas whenever the ambient atmospheric pressure is decreased to a point where the tissue pressure of the gas is more than twice that in the surrounding atmosphere. This results in a marked pressure gradient which drives the gas out of solution. Unlike oxygen and carbon dioxide, which are actively metabolized or transported, the rate of diffusion of the inert gas from the tissues into the expired air is too slow to cope with the volume of inert gas evolved. Thus, the gas comes out of solution locally in the tissues to form bubbles. The number and size of the bubbles in general will be proportional to the diffusion gradient existing between the partial pressure of the inert gas in the tissues and its partial pressure a t the lower ambient pressure. The greater the deviation from the 2: 1 pressure gradient postulated by Boycott et al. (B38), the greater will be the number and size of the bubbles. For a more extensive discussion
112
SHELDON F. GOTTLIEB
of the factors involved in bubble formation the reader is referred to the excellent paper of Harvey (H5). The inert gas generally concerned in hyberbaric medicine is nitrogen. It has an oil-water solubility ratio of approximately 5 :1, which indicates a greater solubility in adipose tissue than in blood. The amount of nitrogen that will dissolve in the tissues depends in part upon the depth and length of the compression as well as the composition of the particular subject. All things being equal, the greater the compression and length of exposure, the greater will be the amount of nitrogen dissolved in the tissues. Obese individuals will dissolve more nitrogen than thin individuals. During decompression the nitrogen will evolve from all the tissues. However, in order to be eliminated, the nitrogen must be transported via the blood. If a large amount of nitrogen has been dissolved, a watery tissue such as blood, capable of containing only about one-fifth as much nitrogen as fat, will not be able to transport in solution the large amount of evolving nitrogen. Superimposed on this is the fact that adipose and bony tissues are poorly vascularieed, thus compounding the perfusion limitation in gas transport. Consequently, intravascular (capillary) and intracellular bubble formation occurs. Recent data indicate that, inter ulia, humoral agents such as bradykinin may have a role in the pathogenesis of dysbarism (C12a). There are various factors that predispose to decompression sickness. Some of these are obesity, age, exercise, and COzaccumulation. In general, older individuals are more susceptible to decompression sickness than younger persons; this is apparently related to the status of the circulatory system. Increased physical activity results in more rapid saturation of the tissues per unit time than in a resting individual. This increased rate of tissue saturation is due to the increased rate of ventilation and circulation resulting in more rapid transport of nitrogen to the tissues. Due to their very rapid circulation and ventilation, small animals are more resistant than man to decompression sickness. Blinks ei! al. (B28) found that increased CO:! tension in the tissues lowers the threshold for bubble formation. Although the mechanism of this phenomenon is not known, there is reason to suspect that it may be due to the high solubility and diffusibility of carbon dioxide (H5). Other factors, such as temperature, water balance, and drugs, may affect the susceptibility to decompression sickness. Results were inconclusive (C24, G27). The symptoms of decompression sickness include bends (local pain), skin rashes, itching, neurological disturbances (scotomata, hemianopsia, diplopia, paresis, paralysis, abnormal reflexes, dysesthesia, aphasia, vertigo), respiratory distress (“chokes”), nausea, and shock (neurocirculatory
HYPERBARIC OXYGENATION
113
collapse). Detailed accounts of the clinical manifestations, predisposing causes, and prevention of decompression sickness have been published recently (D20, L4, N l ) . The principal manifestation of the bends is a deep boring ache or pain in the bones of muscles around the joints. The intensity of the pain can vary. This symptom may be accompanied by neurocirculatory collapse. The cause of the pain is thought to be due to excitation of mechanoreceptors as a result of tissue distortion by the expanding bubbles, or to ischemic excitation due to interference with vascular flow. The role of humoral agents in the production of pain should not be overlooked (K9). Even following the standard U. S. Navy Decompression Tables the incidence of bends may be as high as 4%, depending on age, physical condition, etc. Skin rashes and itching occur fairly often upon decompression. The most common lesion is a mottling of the skin which spreads peripherally. The centers become erythematous and show a slight elevation in temperature. The skin manifestations are found primarily in the pectoral region, back of the shoulders, upper abdomen, forearm, and thigh. The lesions appear to be related to spasm and secondary hyperemia of the small vessels of the skin, apparently caused by bubble formation in the deeper fatty layers of the skin (F5). Gersh and Catchpole (G2) have summarized the CNS symptoms of decompression sickness. The most prominent feature is the occurrence of areas of softening in the spinal cord, usually small and more circumscribed in the white than in the gray matter. Necrosis is more common in the white than in the gray columns. In the white columns, the anterior columns appear to be more susceptible. There is also ascending and descending degeneration of nerve tracts. Hemorrhage is generally not a significant factor in the CNS manifestations of decompression sickness. Neurocirculatory collapse is associated with the more serious cases of CNS involvement in decompression sickness. Another distressing syndrome of dysbarism is “chokes” or respiratory distress. This syndrome is characterized by substernal pain, a nonproductive cough, and dyspnea. There is exacerbation of both the substernal distress and cough when the affected individuals attempt to breathe deeply. To avoid the burning pain the patient may resort to rapid and shallow breathing with subsequent decreases in the arterial PO, and Pco,. The incidence of this syndrome is approximately one-tenth that of bends. It may accompany bends; also it may result in neurocirculatory collapse. For virtually all cases of decompression sickness, save perhaps neurocirculatory collapse which may require oxygenation and blood transfusions, (B41, C19) the most expeditious therapy is recompression. Afflicted indi-
114
SHELDON F. GOTTLIEB
viduals are generally recompressed to 6 ata. The reasoning for compressing to 6 ata is based on the fact that it is the diameter and not the volume of the gas bubble which is the critical dimension determining whether symptoms will occur. It is known from the gas laws that the bubble volume is inversely proportional to the pressure; however, the bubble diameter is proportional to cube root of 27r times the pressure. Compression to 6 ata results in a reduction of the bubble volume to one-sixth of its volume a t 1 atmosphere. At the same time its diameter is reduced by a factor of 0.299; in order to reduce the bubble to one-tenth of its diameter at 1 ata it would be necessary to compress the patient to approximately 150 ata. Compression to 8 ata reduces the bubble diameter only by a factor of 0.27. From these calculations it is obvious that compression beyond 6 ata is of little therapeutic value. One aspect of improper decompression that has been reported is aseptic bone necrosis. Bone is poorly vascularized and poorly innervated; thus, gas bubbles trapped between the periosteum and the bone may lead t o necrosis. Being poorly innervated there is no warning until too late. It is insidious. Bone necrosis has been associated with high pressures, hard work, prolonged exposure, and most important inadequate decompression ! Thus far bone necrosis has been seen only in caisson work. Aseptic bone necrosis at a later date could provide a basis for compensation claims. 3.6. PRECAUTIONS
The protection of medical and ancillary personnel associated with the pressure chamber against the ever present danger of dysbarism makes it mandatory that they be thoroughly grounded in pressure physiology. Mastery of the use of the U. S. Navy or comparable decompression tables is a must, as is knowledge of the various techniques used to facilitate decompression while avoiding the deleterious effects of dysbarism. In addition, all hyperbaric facilities should have an associated chamber with a capability of attaining 6-7 ata. It may even be desirable for the entire hyperbaric facility to have the flexibility of attaining 6-7 ata. Increase in cost of construction for such a flexible facility is less than 1% of the total cost of construction of a hyperbaric chamber capable of only 3 4 ata. The dysbaric experiences of the hyperbaric personnel at Duke University have been discussed (A8). Because of the increased partial pressure of oxygen in the chamber atmosphere there is the remote possibility of an enhancement of combustion. There are reports indicating that the flammability of given materials under variations of pressure remains essentially constant, providing the ratios of the various gases in the system remain constant. Only nonflam-
HYPERBARIC OXYGENATION
115
mable or noncombustible anesthetics should be used. The special problems of anesthesia in a pressure chamber have been reviewed by McDowall (M7), Pittinger (P15), Smith et al. (S25), and Vermeulen-Cranch (V10). All instruments must be explosion-proof and tested for operation under high pressure conditions. Ink-loaded pens will leak upon decompression; vacuum tubes may burst; wrist watches and other instrumentation that will equilibrate with the high pressure may not permit gas to escape as readily upon decompression, with the resultant “gas embolism of instruments.” 4. Analytical Techniques
In recent years immense interest has developed in the use of various analytical techniques for measuring respiratory and anesthetic gases in both the gaseous and dissolved states. The ultimate goal is the determination of gas tensions of specific cells and parts of cells in vivo. A thorough knowledge of the instantaneous gaseous composition of various tissues during health and disease, functional and quiescent states, drug and X-irradiation therapy, anesthesia, etc., would undoubtedly lead to new insight into tissue function, mechanism and prevention of oxygen toxicity, efficacy of chemotherapy, control of respiration, theory of narcosis, etc. The gases of primary concern for measurement are oxygen and carbon dioxide. Although it would be invaluable to have techniques available for measuring instantaneous inert gas tensions in vivo, it is difficult to conceive of a property of the individual gases confering suEicient specificity to be exploited at the present time for the routine differentiation of one of the atmospheric inert gases in the presence of the other atmospheric gases. Specificity is the most important requirement in gas analysis. Techniques dependent on the physical properties of the gas molecules, such as thermal conductivity, density, viscosity, and sound velocity, generally have insufficient specificity to differentiate a single gas in a mixture of gases, and therefore must incorporate in the procedure some type of preliminary separation. Vapor phase fractionation (gas chromatography) is an example of a popular analytical technique based upon a physical property (thermal conductivity) of the gas that requires preliminary separation of the gases by means of special columns (molecular sieve, silica gel, etc.). The detecting and measuring component of a gas chromatography unit contains four heat-sensitive thermistors whose resistances change as a function of temperature; two are used as references and two as detector. The four thermistors form the arms of a Wheatstone bridge. A constant current passing through the bridge heats the thermistors to their effective working range. Thermal equilibrium is attained by passing pure carrier gas
116
SHELDON F. GOTTLIEB
(helium) over the four arms of the bridge. A contaminant (gaseous) in the carrier gas will alter the thermal conductivity properties of the carrier gas. When the carrier gas containing this contaminant comes in contact with a detecting thermistor it alters the rate a t which the thermistor loses heat and thereby causes an imbalance in the bridge. This electrical imbalance is recorded. The degree of change (height of peak or area under the curve) is proportional to the concentration of the contaminant. There are a few techniques depending upon certain physical properties of the gases that can be exploited for analytical measurement. The first is the measurement of oxygen by virtue of its paramagnetic properties (P7). The second technique is based on the absorption and emission spectra of the gases. A substance is paramagnetic if its atoms or molecules possess a permanent magnetic moment. The substance will become magnetized in the Same direction as the applied magnetic field due to the alignment of the individual magnetic moments in this field. Diamagnetic substances do not possess a permanent magnetic moment. For practical purposes oxygen is the only respiratory gas which is paramagnetic and, because of this property, can be measured specifically and accurately in the presence of the other atmospheric gases. Instruments are available commercially for this type of analysis. The absorption and emission spectra of oxygen and nitrogen are well separated but lie in the region of the vacuum ultraviolet. The present cost of this instrumentation for routine gas analysis makes it prohibitive. Carbon dioxide as well as other polyatomic gases (CO, NzO, HzO vapor, organic anesthetic gases) can be measured by nondispersion infrared spectrophotometry. The speed of response and the degree of accuracy obtained by this technique are adequate for routine use during surgery, in clinical cardiac-pulmonary laboratories, and even in many research projects. Nitrogen tension can be measured by ultraviolet emission from ionized nitrogen gas. Nilsson (N3) has recently reviewed the subject of oximetry. The mass spectrometer is theoretically capable of analyzing any gas in a mixture of gases with a speed, specificity, sensitivity, and accuracy that cannot be matched by other known means of analysis. However, the initial cost of the instrument and accessory equipment has thus far precluded its widespread use. Fowler and Hugh Jones described a mass spectrometer specifically for respiratory physiological studies. Extensive use has been made of this instrument (W6-WS). If neither speed of analysis nor continuous analysis is required, the volumetric or manometric techniques may be employed. The volumetric method of gas analysis is based on the selective chemical absorption of components of the gas mixture, first COZand then Oz.The change in volume, a t constant
HYPERBARIC OXYGENATION
117
temperature and pressure, is a measure of the gas concentration (H2). The partial pressures of the gases are calculated from the volume changes, temperature, and ambient pressure, using the gas laws. Manometry involves the measuring of pressure changes while maintaining a constant volume (C7, U1, V9). Ultramicrorespirometry can be employed for studying gas exchange in very small numbers of isolated cells or tissues by use of the Cartesian diver (K7, L15, P5). Manometry can also be used for measuring dissolved gas tensions by first evolving the gases from solution (P11). The techniques discussed briefly above do not permit instantaneous gas analysis of in vivo tissue gas tensions in unanesthetized animals, although some of them may be useful in predicting tissue gas tension. The requirement for a technique capable of recording instantaneous gas tensions in tissues was partially obviated by the introduction of the oxygen cathode by Davies and Brink (Dl). In essence, the principle upon which the oxygen electrode functions is as follows: as the applied voltage across an anode and cathode in a conductive aqueous solution is increased, the current produced from the electrolytic reduction of the oxygen plateaus within the range O.fX.8volt. This implies that at a constant voltage the rate of reduction of the oxygen dissolved in the solution will be directly proportional to the amount of current flowing between the electrodes. The rate of reduction is dependent upon the concentration of the oxygen; thus, a measure of current is a measure of the concentration of dissolved oxygen. All oxygen electrode systems (platinum and silver-silver chloride) operate in this voltage range and show a current flow proportional to the concentration of the oxygen near the electrodes. The oxygen is irreversibly consumed in the measuring process. The early electrode systems were subject to “poisoning” by proteins; in addition, the diffusion geometry was ill-defined. A major breakthrough in the use of oxygen electrodes occurred with the development of a membranecovered electrode by Clark et al. (C16). The membrane protects the electrode surface against poisoning (aging). It also prevents reduction of substances other than oxygen and provides a measure of stability and reproducibility. Excellent discussions of the theory and limitations of the oxygen cathode can be found in the papers of Connelly (C22), Cater et al. (C8), Evans and Naylor (E5), and Severinghaus (A3, S12). The special problems associated with the preparation of the nonpolarizable electrode are discussed by Feder (F3). An extensive source of suppliers of oxygen electrode systems was published recently (M8). The use of the oxygen electrode for measuring oxygen tension under hyperbaric conditions is discussed by Schoemaker (S7). In the same year that Clark modified the oxygen electrode, Stow et al.
118
SHELDON F. GO'ITLIEB
(S33) introduced the C02 electrode. Severinghaus and Bradley modified the COz electrode to increase its sensitivity and response time (Sll). The technique consists essentially of recording the pH of a thin layer of isolated standard solution (KCl and NaHCOa) in contact with the electrode but separated from the tissues by a semipermeable membrane. The membrane is permeable to C02 but impermeable to other ions which might affect the pH of the standard solution. The pH of the solution is completely independent of that of the tissues, Under normal conditions the only gas that influences the pH of the solution is COz. Thus, C02 electrodes provide a pH measurement proportional to the log of the PCO,in the standard solution. The linear relationship of the pH recording to the log of the PCO,of of the tissue. Unlike the oxygen the solution can be affected only by the PCO, electrode which consumes the oxygen in the course of the O2 measurement, the C02 electrode does not consume the COz. Detailed discussions concerning the theory and use of C02 electrodes as well as a list of suppliers have been published recently (Gl, 512). Methods using a combined electrode system have been devised for simultaneous recording of pH, Pcor, and PO*from localized areas of tissue (G18). The electrode systems for in vivo gas tension measurements have the advantage of permitting continuous measurement of gas tensions. However, the methods have certain limitations, especially when the electrode is implanted in specific organs or tissues for in vivo measurements. The insertion of the electrodes into the tissues will provide a mechanical barrier to gas diffusion; it will distort tissues and capillaries contiguous with it and thereby interfere with local blood flow; even capillaries a short distance away may be influenced by the compression; blood vessels may be ruptured and cause edema; local circulatory reflexes may alter the local blood supply; the oxygen electrode by consuming the oxygen will create new diffusion gradients. The measurement of a local oxygen tension is also dependent upon the proximity of the electrode to the capillaries. Electrodes used today measure only mean gas tensions. There is some question whether the oxygen electrode is measuring absolute oxygen tension or just oxygen availability. Advances in electrode technology seem to be revolving around the production of smaller diameter electrodes, new membrane materials, different metals to replace platinum, and improved circuitry. Naturally occurring and synthetic radioactive and nonradioactive isotopes of the atmospheric gases have been employed for studies on various circulatory and respiratory problems as well as their interrelationship. Probably the most commonly employed isotope for studying ventilationperfusion relationships is krypton-85; xenon-133 and oxygen-15 also have been used. Oxygen-15, like the other radioactive gases, can aid in localizing
HYPERBARIC OXYGENATION
119
areas of defective ventilation, but, unlike the other gases, it is metabolized and thereby has the advantage of providing information about regional gas exchange. Because of its short half-life, 2 minutes, oxygen-15 has not been used extensively. Stable isotopes such as oxygen-18 have been used; they suffer the technological disadvantage of requiring more complex instrumentation for measurement than the radioactive gases. 5 . Conclusions
Oxygen is considered a drug, and under certain conditions (vide supra), like all drugs, it is toxic to man. However, its toxic properties should not deter qualified physicians from utilizing this therapeutic approach. Clinical hyperbaric oxygenation is based upon sound physical and physiological principles. In many instances HPOT has proven of immense value. The disappointing results obtained thus far in other clinical areas (coronary infarction, tetanus) should not be permitted to have a negative influence on the future development of this form of therapy; rather they should be an incentive to spur further research in discovering ways of improving the efficacy of HPOT. This involves a fundamental study of the biology, biochemistry, and physiology of oxygen and the other atmospheric gases. Hyperbaric oxygenation has opened up new avenues for basic and clinical research. As more is learned about the physiological and biochemical effects of oxygen and other gases under pressure, it is certain that new uses will be found for this latest addition to the medical armamentarium (B39, G17, M2). One of the basic questions requiring immediate solution for the advancement of hyperbaric oxygenation is concerned with the PO, of tissues during experimentation or therapy. In order to obtain such information, advances must be made in instantaneous ultramicrotechniques for gas tension measurements as well as for simultaneous monitoring of function, metabolism, and respiration on the cellular level. At present, we do not know to what extent ischemic tissues can derive oxygen by diffusion from neighboring areas having normal blood supply. Advances in hyperbaric oxygen therapy of coronary infarctions, neoplasms, and infectious diseases are being hampered by lack of this type of information. As techniques for local gas tension measurement become refined, new information should be forthcoming on the control of the circulation in localized areas in normal and pathological states as well as on the regulation of the circulation in general. New information will be obtained on the effects of altered gas tensions on oxygen metabolism and metabolism in general. New insights will be obtained concerning the PO,required to induce oxygen toxicity. Research into the mechanism of oxygen toxicity must be conducted if
120
SHELDON F. GOTTLIEB
methods are to be devised to prevent its occurrence and thereby eliminate a major obstacle to the full development of hyperbaric oxygenation. Insights obtained from such studies may be useful in determining the causes and treatment of other convulsive disorders. Whether hyperbaric oxygen is a predisposing factor for the action of other stressors has not been investigated. There is a paucity of knowledge concerning the physiological and biochemical compensatory responses resulting from exposure to high pressure oxygen. New information concerning the physics and pharmacology of anesthetic agents, with emphasis on the mechanism of anesthesia, should be forthcoming as a result of increasing the effectiveness of hyperbaric oxygen therapy. The pharmacology of old and new drugs must be extended to include studies on drug reaction in high pressure oxygen and other abnormal gaseous environments. The possibility of synergistic action between chemotherapeutic agents and OHP must be explored further. The synergistic effects of antituberculosis drugs and oxygen (G24) and the enhanced effects of oxygen on the action of fumigants on insects (B34) point out the inability to predict the behavior of a drug in an abnormal gaseous environment. The effect of OHP on the host-parasite relationship has been virtually neglected. The possible use of inert gases as respirable breathing mixtures for hyperbaric environments underscores the need for more basic research into the physiological and biochemical effects of the atmospheric inert gases. Due to the potential fire hazard existing in sealed hyperbaric chambers, new developments in the synthesis of fire-resistant fabrics may be desirable. In addition, new concepts in patient monitoring techniques will undoubtedly result from the peculiar requirements of sealed environmental therapy. The by-product from such studies has wide implications in the fields of space and underwater physiology as well as in clinical and veterinary medicine. In summary, there appears to be cause for enthusiasm concerning the use of hyperbaric oxygenation as a clinical tool. Whether hyperbaric oxygenation becomes a potent therapeutic regimen or fades as a passing fad (albeit an expensive one) depends on its ability to meet the criteria set forth by the Ad Hoc Committee on Hyperbaric Oxygenation of the National Research Council’s Committee on Shock: “It must be uniquely beneficial, having clear-cut advantages over less formidable forms of treatment; it must be fully practical in its use in at least a few important conditions; and, in consequence, it must be applicable to a reasonable number of patients” 641). To help ensure its success, close cooperation will be required between the
HYPERBARIC OXYGENATION
121
basic and clinical research scientists, involved in answering many of the problems discussed above, and the clinician, ultimately concerned with the practical use of high pressure oxygen therapy.
ACKNOWLEDGMENT The author wishes to acknowledge a grant from the Smith, Kline, French Co., Philadelphia, Pennsylvania, which helped in the preparation of this review. The author also wishes to thank Dr. H e m = Ziffer of Roosevelt Hospital in New York, for his critical reading of the manuscript, and Mr. F. Fegan of the Bethlehem Corporation, for supplying detailed information concerning the construction of pressure chambers.
REFERENCES Al. Ad Hoc Committee on Hyperbaric Oxygenation, Committee on Shock, “Hyperbaric Oxygenation: Potentialities and Problems,” pp. 1-37. Natl. Acad. Sci.-Natl. Res. Council, Washington, D.C., 1963. A2. Adam, A., The effects of atmospheres enriched with oxygen upon living organisms, (a) effects upon microorganisms, (b) effects upon mammals experimentally inoculated with tuberculosis, (c) effects upon the lungs of mammals or oxygen pneumonia. Biochem. J . 6, 297-314 (1912). A3. Adam, J. E., and Severinghaus, J. W., Oxygen tension on human cerebral grey and white matter. The effect of forced hyperventilation. J. Neurosurg. 19, 959-963 (1962). A4. Adam, J. F., Ledingham, I. McA., Jackson, J. M., and Smith, G., Combined nitrogen mustard and hyperbaric oxygen therapy in advanced malignant disease. Brit.Med. J . I, 314-315 (1963). A5. Allen, S. C., The role of nitrogen in the problem of oxygen toxicity. Physiologist 4, 2 (1961). A6. Allen, S. C., The comparison of the effects of nitrogen lack and hyperoxia on the vascular development of the chick embryo. Aerospace Med. 34, 897-899 (1963). A7. Allison, M. J., Margolis, G., Chandler, P. J., and Gerszten, E., Effects of hyperbaric oxygen therapy on the development of tuberculosis. Federathn PTOC. 23, 494 (1964). A8. Anderson, B. J., Whalen, R. E., and Saltzman, H. A., Dysbarism among hyperbaric personnel. J . Am. Med. Assoc. 190, 1043-1045 (1964). A9. Antopol, W., Chryssanthou, C., and Kooperstein, S., Effect of air pressure on carcinoma 755 and sarcoma 180 in C57B1 mice. Proc. Am. Assoc. Cancer Res. 3, 205 (1961). A10. Antopol, W. This interpretation was confirmed in personal discussions with Dr. Antopol (1965). A l l . Ashton, N., Effect of oxygen on developing retinal vessels with particular reference to the problem of retrolental fibroplasia. Brit.J . Ophthalmol. 38, 397-432 (1954). A12. Attar, S., Esmond, W. G., and Cowley, R. A., Hyperbaric oxygenation in vascular collapse. J . Thoracic cardiovascuhr Surg. 44, 759-770 (1962). Bl. Back, N., Effect of oxygen tension on the in vitro sensitivity of normal tissues and tumor cells to alkylating agents. PTOC. Am. Assoc. Cancer Res. 3, 207 (1961). B2. Back, N., Ausman, R. K., Weinstein, M., and Lanphier, E. H., Effect of alkylating agents on growth of Vi& jaba bean root tips under various oxygen tensions. FeLratwn PTOC.21, 423 (1962).
122
SHELDON F. GOTTLIEB
B3. Back, N., and Ambrus, J. L., Effect of oxygen tension of normal and tumor tissues to alkylating agents. J . Natl. Cancer Znstil. SO, 17-28 (1963). B4. Barclay, R. S., Ledingham, I. McA., and Norman, J. N., Experimental and human cardiac surgery with hyperbaric oxygen. Zn “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 197-201. Elsevier, Amsterdam, 1964. B5. Barker, J., and Mapson, L. W., Studies in the respiratory and carbohydrate metabolism of plant tissues. VII. Experimental studies with potato tubers of an inhibition of the respiration and of a ‘block’ in the tricarboxylic acid cycle induced by ‘oxygen poisoning.’ PTOC. Roy. SOC.B143, 523-549 (1955). B6. Barker, J., Quartley, C. E., and Turner, E. R., Studies in the respiratory and carbohydrate metabolism of plant tissues. IX. Experimental studies of the influence of oxygen a t high pressures on the respiration of apples and of a ‘block’ in the Roy. SOC.B162,88-108 tricarboxylic acid cycle induced by ‘oxygen poisoning.’ PTOC. (1960). B7. Barker, J., Studies in the respiratory and carbohydrate metabolism of plant tissues. X. The influence of oxygen a t high pressures as a stimulant and inhibitor of certain pathways of respiration in carrots. PTOC.Roy. Soc. B164, 289-308 (1961). B8. Barrie, H., Hyperbaric oxygen in resuscitation of the newborn. Lancet 11, 1223-1224 (1963). B9. Bean, J. W., and Siegfried, E. C., Transient and permanent after-effects of exposure to oxygen a t high pressure. Am. J. Physwl. 143, 656-665 (1945). B10. Bean, J. W., Wagner, S., and Siegfried, E. C., Residual disturbances in the higher functions of the C. N. 8. induced by oxygen a t high pressure. Am. J. Physiol. 143, 206-213 (1945). B11. Bean, J. W., Effects of oxygen a t increased pressures. Physwl. Rev. 26, 1-147 (1945). B12. Bean, J. W., The hypophysis as a determinant in the reaction of the mammal to oxygen a t high pressure. Am. J . Physiol. 170, 508-517 (1952). B13. Bean, J. W., and Smith, C. W., Hypophyseal and adrenocortical factors in pulmonary damage induced by oxygen a t atmoBpheric pressure. Am. J. Physiol. 172, 169-179 (1953). B14. Bean, J. W., and Johnson, R. C., Epinephrine and neurogenic factors in the 2 a t high pressure. Am. J. pulmonary edema and CNS reactions induced by 0 Physiol. 180, 438-444 (1955). B15. Bean, J. W., Tris buffer, Cog and sympathetic adrenal system in reaction to On a t high pressure. Am. J. Physiol. 201, 737-739 (1961). B16. Bean, J. W., Brain pol in exposures to 02 a t high pressure (OHP). Federation PTOC. 20, 100 (1961). B17. Becker, N. H., and Galvin, J. F., Effect of oxygen-rich atmospheres on cerebral lipid peroxides. Aerospace Med. 33, 985-987 (1962). B18. Beehler, C. C., Newton, N. L., Culvey, J. L., and Tredici, T., Ocular hyperoxia. Aerospace Med. 34, 1017-1020 (1963). B19. Behnke, A. R., h w , L. A., Shilling, C. W., Thomson, R. M., and Messer, A. C., Studies on the effects of high oxygen pressure. I. Effect of high oxygen pressure upon the carbondioxide and oxygen content, the acidity and the carbon-dioxide combining power of the blood. Am. J. Physiol. 107, 13-28 (1934). B20. Behnke, A. R., Jr.,.Thomson, R. M., and Motley, E. P., The psychologic effects from breathing air a t 4 atmospheres pressure. Am. J. Physwl. 112,554-558 (1935).
HYPERBARIC OXYGENATION
123
B21. Behnke, A. R., Jr., and Yarbrough, 0. D., Respiratory resistance, oil water solubility and mental effects of argon, compared with helium and nitrogen. Am. J . Physwl. 126,409415 (1939). B22. Behnke, A. R., Jr., Physiologic studies pertaining to deep sea diving and aviation especially in relation to the fat content and composition of the body. Harvey Lectures 57, 198-226 (1941). B23. Berfenatam, B., Edlund, T., and Zettergren, L., The hyaline membrane disease, a review of earlier clinical and experimental findings and some studies on the pathogenesis of hyaline membranes in 02-intoxicated rabbits. A c b Paediat. 47, 82-100 (1958). B24. Berfenatam, B., Edlund, T., and Zettergren, L., Hyaline membrane disease. The influence of high oxygen concentration on ciliary activity in the respiratory tract. An experimental study on rabbits. Acla Paediat. 47, 527-533 (1958). B25. Bernhard, W. F., Somers, L. A., Kriek, H. R., and Tadaaki, A., Corrective and palliative surgery in infants and children with congenital heart disease. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 189-193. Elsevier, Amsterdam, 1964. B26. Bert, P., “Barometric Pressure.” College Book Co., Columbus, Ohio, 1943; (translation of “La pression baromktrique.” Masson, Paris, 1878). B27. Blair, E., Henning, G., Esmond, W. G., Attar, S., Conley, R. A., and Michaels, M., The effect of hyperbaric oxygenation (OHP) on three forms of shock-traumatic, hemorrhagic and septic. J . Trauma 4, 652-663 (1964). B28. Blinks, L. R., Twitty, V. C., and Whitaker, D. M., Bubble formation in frogs and rats. In “Manual of Decompression Sickness” (J. F. Fulton, ed.), pp. 152-159. Saunders, Philadelphia, Pennsylvania, 1951. B29. Bloor, K., Bratten, N. T., Jacobson, I., McCaffrey, J. F., and McDowall, D. G., Low flow perfusion with total heart-lung bypass a t 2 atmospheres absolute pressure. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 194-196. Elsevier, Amsterdam, 1964. B30. Boerema, I., Kroll, J. A., Meijne, N. G., Lokin, E., Kroon, B., and Huiskos, J. W., High atmospheric pressure as an aid to cardiac surgery. Arch. Chir. Need 8,193-211 (1956). B31. Boerema, I., and Brummelkamp, W. H., Treatment of anerobic infections by the introduction of pure oxygen under a pressure of three atmospheres. Ned. Tijdschr. Geneesk. 104, 2548-2550 (1960). B32. Boerema, I., An operating room with high atmospheric pressure. Surgery 47, 291-298 (1961). B33. Bond, E.J., Effect of oxygen on cyanide poisoning in insects. Nature 195, 10021003 (1962). B34. Bond, E. J., The action of fumigants on insects. IV. The effects of oxygen on the toxicity of fumigants to insects. Can. J . Biochem. Physwl.41, 993-1004 (1963). B35. Bondurant, S., and Smith, C., Effects of oxygen intoxication on the surface characteristics of lung extracts. Physiologist 6, 111 (1962). B36. Boothby, W. M., Mayo, C. W., and Lovelace, W. R., One hundred per cent oxygen-indications for its use and methods of its administration. J . Am. Med. Assoc. 115,477482 (1939). B37. Boriskin, V. V., Oblapenko, P. V., Rol’nile, V. V., and Sabin, M. M., Development potentialities of the animal organism when atmospheric nitrogen is replaced by helium. Dokl. Akud. Nauk SSSR 149, 475478 (1962).
124
SHELDON F. GOTTLIEB
B38. Boycott, A. E., Damant, G. C. C., and Haldane, J. S., Preventation of compressedair illness. J. Hyg. 8, 342-343 (1908). B39. Brauer, R. W., Liver circulation and function. Physwl. Rev. 45, 115-213 (1963). B40. Bruce, A. K., and Stannard, J. N., The effect of x-irradiation upon potassium retentivity of yeast. J. Cellular Comp. Physwl. 61,325-340 (1958). B41. Brunner, F. P., Frick, P. G., and Bmuhlman, A. A., Post-decompression shock due to extravasation of plasma. Lancet I, 1071-1073 (1964). B42. Brummelkamp, W. H., Hoogendijk, J. L., and Boerema, I., Treatment of anerobic infections (clostridial myositis) by drenching the tissues with oxygen under high atmospheric pressure. Surgery 49, 299-302 (1961). B43. Brummelkamp, W. H., Boerema, I., and Hoogendijk, J. L., Treatment of clostridial infections with hyperbaric oxygen drenching. Lancet 11, 235-238 (1963). B44. Brummelkamp, W. H., Treatment of infectionswith Clostridium vrelchii by oxygen therapy a t 3 atmospheres. A report on 37 cases. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 20-30. Elsevier, Amsterdam, 1964. B45. Btummelkamp, W. H., Hyperbaric oxygen therapy in tetanus. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 63-67. Elsevier, Amsterdam, 1964. B46. Bureau of Investigation., “The Cunningham ‘tank treatment’.’’ J. Am. Med. Assoc. 90, 1494-1496 (1928). B47. Burnett, W., Clark, R. G., Duthrie, H. L., and Smith, A. N., The treatment of shock by oxygen under pressure. Scot. Med. J. 4, 535-538 (1959). Cl. Cameron, A. J. V., Gibb, B. H., Ledingham, I. McA., McGuiness, J. B., Norman, J. N., and Sharif, M., A controlled clinical trial of hyperbaric oxygen in the treatment of acute myocardial infarction. Preliminary results. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 75-83. Elsevier, Amsterdam, 1964. C2. Campbell, J. A., Body temperature and oxygen poisoning. J. Physwl. (London) 89, 17P-18P (1937). C3. Campbell, J. A,, Oxygen poisoning and tumor growth. Brit. J. Ezptl. Pathol. 18, 191-197 (1937). C4. Campbell, K.,Intensive oxygen therapy as a possible cause of retrolental fibroplasia, a clinical approach. Med. J. Australia 38,48-50 (1951). C5. Carpenter, F. G., Depressant action of inert gases on central nervous system in mice. Am. J. Physiol. 173, 471474 (1953). C6. Carpenter, F. G., Alteration in mammalian nerve metabolism by soluble and gaseous anesthetics. Am. J. Physiol. 187, 573-578 (1956). C7. Cassin, S,, and Vogh, B., Automatic small animal volumeter. J. Appl. Physwl. 17, 150-151 (1961). C8. Cater, D. B., Silver, I. A., and Wilson, G. M., Apparatus and technique for the quantitative measurement of oxygen tension in living tissues. Proc. Roy. SOC. B161, 256-276 (1959). C9. Cater, D. B., and Silver, I. A., Quantitative measurement on oxygen tension in normal tissues and in tumors of patients before and after radiotherapy. Acta Radiol. 63, 233-256 (1960). C10. Cater, D. B., Schoeniger, E. L., and Watkinson, D. A., Effect on oxygen tension of tumours of breathing oxygen a t high pressures. Lancet 11, 7252 (1962). C11. Cedergren, B., Gyllenstein, L., and Wereall, J., Pulmonary damage caused by
HYPERBARIC OXYGENATION
125
oxygen poisoning: an electron microscopic study on mice. A c h Paediat. 48, 477-494 (1959). C12. Chapin, J. L., Anticonvulsant threshold of COz in oxygen under high pressure. Proc. Sac. Exptl. Bwl. Med. 90, 663-664 (1960). C12a. Chryseanthou, C. Kalberer, J., Jr., Kooperstein, S., and Antopol, W., Studies on dysbarism. 11. Influence of bradykinin and “bradykinin-antagonists” on decompression sickness in mice. Aerospace Med. 36,741-746 (1964). C13. Churchill-Davidson, I., Sanger, C., and Thomlinson, R. H., High pressure oxygen and radiotherapy. Lancet I, 1091-1095 (1955). C14. Churchill-Davidson, I., Sanger, C., and Thomlinson, R. H., Oxygenation in radiotherapy. 11. Clinical application. Brit. J. Radiol. SO, 406422 (1957). C15. Churchill-Davidson, I., The use and effects of high-pressure oxygen in radiotherapy. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 140-143. Elsevier, Amsterdam, 1964. C16. Clark, L. C., Jr., Wolf, R., Granger, D., and Taylor, Z., Continuous recording of blood oxygen tensions by polarography. J . Appl. Physiol. 6 , 189-193 (1953). C17. Clements, J. A., Surface phenomena in relation to pulmonary function. Physiologist 6, 11-28 (1962). C18. Cleveland, L. R., Toxicity of oxygen for protozoa in viva and in vitro: animals defaunated without injury. Biol. Bull. 48, 455-468 (1925). C18a. Cochran, W. D., Levinson, H., Muirhead, D. M., Jr., Wesley, R., Wang C. S., and Smith, C. A., A clinical trial of high oxygen pressure for the respiratory digtress syndrome. New Engl. J . Med. 272, 347-351 (1965). C19. Cockett, A. T. K., and Nakamura, R. M., A new concept in the treatment of decompression sickness (Dysbarism). Lancet I, 1102 (1964). (220. Collier, C. R., Alterations of surfactant in oxygen poisoning. Proc. 30th Ann. Meeting, Am. Coll. Chest Physicians p. 50 (1964). C21. Comroe, J. H., Jr., Drips, R. D., Dumke, P. R., and Deming, M., Oxygen toxicity. The effect of inhalation of high concentrations of oxygen for twenty-four hours on normal men a t sea level and a t a simulated altitude of 18,000 ft. J . Am. Med. Assoc. 128, 710-717 (1945). C21a. Conger, A. D., and Fairchild, L. M., Breakage of chromosomes by oxygen. Proc. Natl. Acad. Sci. U.S. 98, 289-299 (1952). C22. Connelly, C. M., Methods for measuring tissue oxygen tension; theory and evaluation: the oxygen electrode. Federation Proc. 16, 681-684 (1957). C23. Cook, S. F., Effect of helium and argon on metabolism and metamorphosis. J. Cellular Comp. Physiol. 36, 115-127 (1950). C24. Cook, S. F., Role of exercise, temperature, drugs, and water balance in decompression sickness. In “Manual of Decompression Sickness.” (J. F. Fulton, ed.), pp. 235-241. Saunders, Philadelphia, Pennsylvania, 1951. C25. Cook, S. F., South, F. E., Jr., and Young, D. R., Effect of helium on gas exchange in mice. Am. J . Physwl. 164, 248-250 (1951). C26. Cook, S. F., and South, F. E., Jr., Helium and comparative in vitro metabolism of mouse-tissue slices. Am. J . Physwl. 173, 542-544 (1953). C27. Cope, C., The importance of oxygen in the treatment of cyanide poisoning. J. Am. Med. Assoc. 176, 1061-1064 (1961). C28. Cowley, R. A., Attar, S., Esmond, W. G., Blair, E., and Hawthorne, I., Electrocardiographic and biochemical study in hemorrhagic shock in dogs treated with hyperbaric oxygenation. Circulation 27, 87-75 (1963).
126
SHELDON F. GOTTLIEB
C29. Cowley, R. A., Attar, S., Esmond, W. G., and Blair, E., The utilization of hyperbaric oxygenation on hemorrhagic shock in dogs. I n “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 177-181. Elsevier, Amsterdam, 1964. C30. Crawford, J. S., Hyperbaric oxygen in resuscitation of the newborn. Lancet 11, 815-816 (1964). C31. Cross, K. W., Dawes, G. S., Hyman, A., and Mott, J. C., Hyperbaric oxygen and intermittent positive-pressure ventilation in resuscitation. Lancet II, 560-562 (1964). C32. Cullen, S. C., and Gross, E. G., Anesthetic properties of xenon in animals and human beings with some additional observations on krypton. Science 113, 580-582 (1951). C33. Cunningham, 0. J., Oxygen therapy by means of compressed air. Anesthesia Analgesia 6, 64-66 (1927). D1. Davies, P. W., and Brink, F., Microelectrodw for measuring local oxygen tension T .524-533 (1942). in animal tissues. Rev. Sci. I ~ s ~13, D2. Davis, J. A., and Tizard, J. P. M., Hyperbaric oxygen in resuscitation of the newborn. Lancet I, 166 (1964). D3. Dawes, G. S., and Cross, K. W., Hyperbaric oxygen in resuscitation of the newborn. Lancet 11, 910 (1964). D4. De Almeida, A. O., Researches BUT l’action toxique des hautes pressions d’oxyghe. Compt. Rend. SOC.Biol. 116, 1225-1227 (1934). D5. De Almeida, A. O., Traitement et gu6rison, par l’oxyghne, du cancer expMmenta1 des rats. Compt. Rend. SOC.BWZ. 116, 1228-1230 (1934). D6. De Almeida, A. O., and Pocheco, G., Emaim de tratamento das gangranas gazosas experimentais pel0 oxigenio em attar pressoea e pel0 oxigenio em eatado nascente. Rev. B r a d . BWZ. 1, 1-10 (1941). D7. De Come, J. J., and Rogers, L. S., Effect of hyperbaric oxygen and cancer chemotherapy on growth of animal tumors. Burg. Forum 10, 203-205 (1964). D8. Dickens, F., The toxic effects of oxygen on brain metabolism and on tissue enzymes. 1. Brain metabolism. Biochem. J. 40, 145-171 (1946). D9. Dickens, F., The toxic effects of oxygen on brain metabolism and on tissue enzymes. Biochem. J. 40, 171-187 (1946). D10. Di Paolo, J. A., Influence of oxygen concentration on formation of lung tumors by transplacental exposure to urethan. Federation PTOC.20, 154 (1961). D11. Di Paolo, J. A., Effects of oxygen concentration on carcinogenwis induced by transplacental exposure to urethan. Cancer Res. 22, 299-304 (1962). D12. Dixon, M., and Webb, E. C., “Enzymes,” 2nd ed. Academic Prws, New York, 1964. D13. Dollery, C. T., Hill, D. W., Mailer, C. M., and Ramalho, P. S., High oxygen pressure and the retinal blood vessels. Lancet 11, 291-292 (1964). D14. Domino, E. F., Gottlieb, S. F., Brauer, R. W., Cullen, S. C., and Feathemtone, R. M., Effects of xenon a t elevated pressures in the dog. Anesthesiology 26, 43-53 (1964). D15. Donald, I., Hyperbaric oxygen in the resuscitation of the newborn. Lancet 11, 1383 (1963). D16. Donald, K. W., Oxygen poisoning in man. Brit. Med. J. I, 667-672 (1947). D17. Donald, K. W., Oxygen poisoning in man. Brit. Med. J. I, 712-717 (1947). D18. Douglas, T. A., Lawson, D. D., Ladingham, I. McA., Norman, J. N., Sharp,
HYPERBARIC OXYGENATION
127
G. R., and Smith, G., Carbon monoxide poisoning. A comparison between the efficiencies of oxygen a t one atmosphere pressure, of oxygen a t two atmospheres pressure, and of 5% and 7% carbon monoxide in oxygen. Lancet I, 68-69 (1962). D19. Douglas, T. A,, Lawson, D. D., Ledingham, I. McA., Norman, J. N., Sharp, G. R., and Smith, G., Carbon monoxide poisoning. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 161-165. Elsevier, Amsterdam, 1964. D20. Duffner, G. J., Scuba diving injuries, predisposing causes and prevention. J. Am. Med. Assoc. 176, 375-378 (1961). D21. Du &dt, L. A., The effect of oxygen on the response of spontaneous tumours in mice to radiotherapy. Brit. J. Radiol. 36, 749-754 (1963). D22. Du Sault, L. A., Optimum Oz/COzratio in radiotherapy. Radiology 82,333 (1964). El. Ebert, M., Homey, S., and Howard, A. N., Effect of inert gases on oxygen-dependent radiosensitivity. Nature 181, 613-616 (1958). E2. Emery, E. W., Lucas, B. G. B., Williams, K. G., Technique of irradiation of conscious patients under increased oxygen pressure. Lancet I, 248-250 (1960). E3. End, E., and Long, C. W., Oxygen under pressure in carbon monoxide poisoning. I. Effect on dogs and guinea pigs. J. Ind. Hyg. Toxicol. 24, 302-306 (1942). E4. Esmond, W. G., Attar, S., and Cowley, R. A., Hyperbaric oxygenation in experimental hemorrhagic shock: experimental chamber design and operation. Trans. Am. SOC.Artificial Internal Organs 8, 384-391 (1962). E5. Evans, N. T. S. and Naylor, P. F. D., The measurement of partial pressure of oxygen in vivo. J. Polarog. SOC.2 , 2-8 (1960). E6. Evans, N. T. S., and Naylor, P. F. D., The effect of oxygen breathing and radiotherapy upon the tissue oxygen tension of some human tumours. Brit. J. Radiol. 86, 418425 (1963). E7. Evans, W. E., Darin, J. C., End, E., and Ellison, E. H., The use of hyperbaric oxygen in the treatment of endotoxin shock. Surgery 66, 184-192 (1964). F1. Falaetti, H., Effect of oxygen tension on sodium transport across isolated frog skin. Proc. SOC.Exptl. Bwl. Med. 101, 721-722 (1959). F2. Featherstone, R. M., and Muehlbaecher, C. A., The current role of the inert gases in the search for anesthesia mechanisms. Pharmucol. Rev. 16, 97-121 (1963). F3. Feder, W., Silver-silver chloride electrode as a nonpolarizable bioelectrode. J . Appl. Physwl. 18, 397-401 (1963). F4. Ferm, V. H., Teratogenic effects of hyperbaric oxygen. Proc. SOC.Exptl. Biol. Med. 116, 975-976 (1964). F5. Ferris, E. B., and Engel, G. L., Clinical nature of high altitude decompression sickness. I n “Manual of Decompression Sickness” (J. F. Fulton, ed.), pp. 35-37. Saunders, Philadelphia, Pennsylvania, 1951. F6. Foster, C. A., and Churchill-Davidson, I., Response to high pressure oxygen of conscious volunteers and patients. J. Appl. Physiol. 18, 492-496 (1963). F7. Frank, H. A,, and Fine, J., Traumatic shock: a study of the effect of oxygen on hemorrhagic shock. J. Clin. Invest. 22, 305-314 (1943). F8. Frankel, J., and Schneiderman, H. A., The effects of nitrogen, helium, argon, and sulfur hexafluoride on the development of insects. J. Cellular Comp. Physiol. 02, 431457 (1958). F9. Frankenhaeusser, M., Graff-Lennevig, V., and Hesser, C. M., Psychomotor performance in man as affected by high oxygen pressure (3 atmospheres). Acta Physiol. Smnd. 60, 1-7 (1960).
128
SHELDON F. GOTTLIEB
G1. Gambino, S. R., Determination of blood Pco,. Clin. Chem. 7 , 236-245 (1961). G2. Gersh, I., and Catchpole, H. R., Decompression sickness, physical factors and pathological consequences. In “Manual of Decompression Sickness” (J. F. Fulton, ed.), pp. 161-181. Saunders, Philadelphia, Pennsylvania, 1951. G3. Gerschman, R., and Fenn, W. O., Ascorbic acid content of adrenal glands of rat in oxygen poisoning. Am. J . Physiol. 176, 6-8 (1954). G4. Gerschman, R., Gilbert, D. L., and Caccamise, D., Effect of various substances on survival times of mice exposed to different high oxygen tensions. Am. J. Physiol. 192, 563-671 (1958). G5. Gerschman, R., Gilbert, D. L., Nye, G. W., Dwyer, P., and Fenn, W. O., Oxygen poisoning and x-irradiation: a mechanism in common. Science 119, 623-626 (1954). G6. Gerschman, R., Arguellas, A. E., and Ibeas, M. M., Effects of high oxygen tensions on mammalian gonads. Proc. bdnd Intern. Congr. Intern. Union Physiol. Sci., Leiden, 196.8. Vol. 2, p. 357 (Abstract). Excerpta Med. Found., Amsterdam, 1962. G7. Gesell, R., On the chemical regulation of respiration. 1. The regulation of respiration with special reference t o the metabolism of the respiratory center and the coordination of the dual function of hemoglobin. Am. J . Physiol. 66, 5-49 (1923). G7a. Gilbert, D. L., Gerschman, R., Cohen, J., and Sherwood, W., Influence of high oxygen pressure on viscosity of solutions of sodium deoxyribonucleic acid and sodium alginate. J . Am. Chem. SOC.79, 5677-5680 (1957). G8. Gilbert, D. L., Gerschman, R., Fenn, W. O., and Dwyer, P., Effect of fasting and x-irradiation on oxygen poisoning in mice. Am. J . Physiol. 181, 272-274 (1955). G9. Gilbert, D. L., The role of prooxidants and antioxidants in oxygen toxicity. Radiation Res. Suppl. 3, 44-53 (1963). G10. Gilbert, D. L., and Lowenberg, W. E., Influence of high oxygen pressure on the resting membrane potential of frog sartorius muscle. J . Cellular Comp. Physiol. 64, 271-278 (1964). G11. Glinka, Z., and Reinhold, L., Rapid changes in permeability of cell membranes to water brought about by carbon dioxide and oxygen. Plant Physiol. 37, 481-486 (1962). G12. Glover, J. L., and Mendelson, J., Effects of hyperbaric oxygenation on rabbits with Clostridium perfringens infections. J. Trauma 4, 624-651 (1964). G13. Goldfeder, A., and Clarke, G. E., The response of neoplasms to x-irradiation in vivo a t increased oxygen tensions. Radiation Res. 13, 751-767 (1960). G14. Goodlin, R. C., Foetal incubator. Lancet I, 1356-1357 (1962). G15. Goodlin, R. C., Hyperbaric resuscitation of the newborn. Lancet I, 382 (1964). G16. Gordon, E. R., Tinker, 0. O., and Watson, W. J., A study of intermediate carbohydrate metabolism in rats exposed to high oxygen pressures. Can. J . Biochem. Physiol. 41, 327-333 (1963). G17. Gorecki, Z., Oxygen under pressure applied directly to bedsores: case report. J. Am. Geriat. SOC.12, 1147-1148 (1964). G18. Gotoh, F., and Meyer, J. S., A combined electrode for recording absolute tensions of oxygen and carbon dioxide from small areas of tissue. Electroencephalog. Clin. Neurophysiol. 13, 119-122 (1961). G19. Gottlieb, S. F., and Jagodzinski, R. V., A comparison of xenon, ether, and nembutal anesthesia in the rat. Physiologist 6 , 149 (1962). G20. Gottlieb, S. F., The possible use of high pressure oxygen in the treatment of leprosy and tuberculosis. Diseases Chest 44,215-217 (1963).
HYPERBARIC OXYQENATION
129
G21. Gottlieb, S. F., and Jagodzinski, R. V., Use of microbes for the cellular study of oxygen toxicity. Federation Proc. 21, 226 (1962). G22. Gottlieb, S. F., and Jagodzinski, R. V., Role of THAM in protecting mice against convulsive episodes caused by exposure to oxygen under high pressure. PTOC.SOC. ExpU. Biol. Med. 112, 427430 (1963). G23. Gottlieb, S. F., Rose, N. R., Maurizi, J., and Lanphier, E. H., Inhibitory effects of hyperbaric oxygen on bacteria and fungi. Lancet I, 382 (1964). G24. Gottlieb, S. F., Rose, N. R., Maurizi, J., and Lanphier, E. H., Oxygen inhibition of growth of Mycobacterium tuberculosis. J. Bacteriol. 87, 838-843 (1964). G25. Gottlieb, S. F., Hyperbaric oxygen in resuscitation of the newborn. Lancet I, 220-221 (1965). G26. Gottlieb, S. F., and Weatherly, J. M., Physiological effects of the noble gases on the frog sciatic nerve and gastrocnemius muscle. Am. J. Physiol. 208, 407-411 (1965). G27. Gray, J. S., Constitutional factors affecting susceptibility t o decompression sickness. I n “Manual of Decompression Sickness” (J. F. Fulton, ed.), pp. 182-191. Saunders, Philadelphia, Pennsylvania, 1951. G28. Gray, L. H., Oxygenation in radiotherapy. I. Radiobiological considerations. Brit. J. Radiol. 30, 403-406 (1957). G29. Gray, L. H., Conger, A. D., Ebert, M., Hornsey, S., and Scott, 0. C. A., The concentration of dissolved oxygen in tissues a t the time of irradiation as a factor in radiotherapy. Brit. J. Radiol. 26, 638-651 (1953). G30. Grossman, M. S,, and Penrod, K. E., The thyroid and high oxygen poisoning in rats. Am. J. Physiol. 186, 182-184 (1949). G31. Gyllenstein, L. J., and Hellstrom, B. E., Experimental approach to retrolental fibroplasia. Am. J . Ophthalmol. 39, 475-488, (1955). H1. Haldane, J. S., The relation of the action of carbonic oxide t o oxygen tension. J . PhysWl. (London) 18,201-217 (1895). H2. Haldane, J. S., and Graham, J. I., “Methods of Air Analysis,” 4th ed. Griffin, London, 1935. H3. Hale, H. B., Williams, E. W., Anderson, J. E., and Ellis, J. P., Jr., Endocrine and metabolic effects of shortduration hyperoxia. Aerospace Med. 36, 449-451 (1964). H4. Harris, R. H., and Hitchcock, C. R., Serum transaminase levels (SGOT) in dogs with induced myocardial infarction treated with hyperbaric oxygenation. I n “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 110-115. Elsevier, Amsterdam, 1964. H5. Harvey, E. N., Bubble formation. Natl. Acad. Sci.-Natl. Res. Council Publ. 377, 53-60 (1955). H6. Haugaard, N., Oxygen poisoning. The relation between inactivation of enzymes by oxygen and essential sulfhydryl groups. J. BWZ. Chem. 164, 265270 (1946). H7. Haugaard, N., Effect of high oxygen tensions upon enzymes. Natl. Acad. Sci.NaU. Res. Council Publ. 377, 8-12 (1955). H7a. Haugaard, N., The toxic action of oxygen on metabolism and the role of trace metals. I n ‘‘Oxygen in the Animal Organism” (F. Dickens and E. Neil, eds.), pp. 495-507. Pergamon Press, New York, 1964. H8. Heston, W. E., and Pratt, A. W., Increase in induced pulmonary tumors in mice associated with exposure to high concentrations of oxygen. PTOC.SOL Exptl. Bwl. Med. 92, 451454 (1956).
130
SHELDON F. GOTTLIEB
H9. Heston, W. E., and Pratt, A. W., Effect of concentration of oxygen in occurrence of pulmonary tumors in strain A mice. J. Natl. Cancer Inst. 22, 707-717 (1959). H10. Hmton, D., A method for the arrest of spreading gas gangrene by oxygen injection. Am. J. Surg. 78, 228-232 (1947). H11. Hitchcock, C. R., Experimental pressurized operating room. Congressional Record March, p. 274 (1962). H12. Hoge, S. F., Oxygen in gas infection. J . Arkansas Med. SOC.28, 4-9 (1932). H13. Hopkinson, W. I., and Towers, A. G., Effects of hyperbaric oxygen on some common pathogenic bacteria. Lancet 11, 1361-1363 (1963). H14. Howard-Flanders, P., and Wright, E. A., Effect of oxygen on the radiosensitivity of growing bone and a possible danger in the use of oxygen during radiotherapy. Nature 176, 428-429 (1955). H15. Hulpieu, H. R., and Cole, V. V., The effect of humidity and temperature on oxygen toxicity. J . Lab. Clin. Med. 29, 1134-1138 (1944). H16. Hunter, 5. W., Long, 8. V., Berger, E. C, and Britton, D., In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 105-109. Elsevier, Amsterdam, 1964. H17. Hutchison, J. H., Kerr, M. M., McPhail, M. F. M., Douglas, T. A., Smith, G., Norman, J. N., and Bates, E. H., Studies in the treatment of the pulmonary syndrome of the newborn. Lancet 11, 465-469 (1962). H18. Hutchison, J. H., Kerr, M. M., Williams, K. G., and Hopkinson, W. I., Hyperbaric oxygen in the resuscitation of the newborn. Lancet 11, 1019-1022 (1963). H19. Hutchison, J. H., Kerr, M. M., Williams, K. G., and Hopkinson, W. I., Hyperbaric oxygen in the Tesuscitation of the newborn. Lancet 11, 1334-1335 (1963). H20. Hutchison, J. H., Kerr, M. M., Williams, K. G., and Hopkinson, W. I., Hyperbaric oxygen in resuscitation of the newborn. Lancet 11, 691-692 (1964). 11. Illingworth, C. F. W., Smith, G., Lawson, D. D., Ledmgham, I. McA., Sharp, G. R., and Griffiths, J. C., Surgical and physiological observations in an experimental pressure chamber. Brit. J . Surg. 49, 222-227 (1961). 12. Ivanov, K. P., Effect of increased oxygen pressure in animals poisoned by potassium (1959). cyanide. Farmakol. Toksikol. 22, 46-73 J1. Jacobson, I., and Lawson, D. D., The effect of hyperbaric oxygen on experimental cerebral infarction in the dog with preliminary correlations of cerebral blood flow a t 2 atmospheres of oxygen. J. Neurosurg. 20, 849-859 (1963). 52. Jacobson, I., Harper, A. M., and McDowall, D. G., The effects of oxygen at 1 and 2 atmospheres on the blood flow and oxygen uptake of the cerebral cortex. Surg. Gyneco2. Obstet. 119, 737-747 (1964). J3. Jacobson, J. H., Morsch, J. H. C., and Rendell-Baker, L., The historical perspective of hyperbaric therapy. I n “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummekamp, and N. G. Meijne, eds.), pp. 7-19. Elsevier, Amsterdam, 1964. 54. Jamieson, D., and Van den Brenk, H. A. S., Pulmonary damage due to high pressure oxygen breathing in rats. 2. Changes in dehydrogenase activity of rat lung. Australian J . Exptl. Biol. Med. Sci. 42, 51-56 (1962). 55. Johnson, F. H., and Lewin, E., The disinfection of E. coli in relation to temperature, hydrostatic pressure and quinine. J. Cellular Comp. Physwl.28, 2345 (1946). 56. Johnson, F. H., Eyring, H., and Polbar, M. J., ede., Hydrostatic pressure and molecular volume changes. “The Kinetic Basis of Molecular Biology,” pp. 286368. Wdey, New York, 1954.
HYPERBARIC OXYGENATION
131
K1. Karasevich, E. G., Harper, E. M., Sharp, N. C. C., Shields, R. S., Smith, G., and McDowall, D. G., Hyperbaric oxygen in clostridial infections. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 36-40. Elsevier, Amsterdam, 1964. K2. Karsner, H. T., The pathological effects of atmospheres rich in oxygen. J . Exptl. Med. 23, 149-169 (1916). K3. Karsner, H. T., Brittingham, H. H., Richardson, M. L., Influence of high partial pressures of oxygen upon bacterial cultures. J . Med. Res. 44, 83-88 (1923). K4. Karsner, H. T., and Saphir, O., Influence of high partial pressures of oxygen on the growth of certain molds. J . Infect. Diseases 59, 231-236 (1926). K5. Kaufman, €3. D., Owen, S. G., and Lambertsen, C. J., Effects of brief interruptions of pure oxygen breathing upon central nervous system tolerance to oxygen. Federation Proc. 16, 107 (1956). K6. Kelley, H. G., and Page, W. G., Treatment of anaerobic infections in mice with hyperpressure oxygen. Surg. Forum 14, 46-47 (1963). K7. Kieler, J., Cultivation of leukemic cells in the Cartesian diver. In “The Leukemias” (J. W. Rebuck, F. H. Bethell, and R. W. Monto, eds.), pp. 215-226. Academic Press, New York, 1957. K8. Killick, E. M., and Marchant, J. V., Resuscitation of dogs from severe acute carbon monoxide poisoning. J . Physiol. (London) 147, 274-298 (1959). K9. Kim, R. K. S., in Discussion following the paper of Horton, E. W., The role of bradykinin in the peripheral nervous system. Ann. N . Y . Acad. Sci. 104, 250-257 (1963). K10. Kinsey, D. L., Hyperbaric oxygen and 5-fluorouracil in the treatment of experimental melanoma. Surg. Forum. 16, 205-206 (1964). K l l . Kluft, O., and Boerema, I., Hyperbaric oxygen in experimental cancer in mice. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 126-136. Elsevier, Amsterdam, 1964. K12. Krernentz, E. T., and Knudson, L., The effect of increased oxygen tension on the turnorcidal effect of nitrogen mustard. Surgery 60, 266-273 (1961). K13. Kydd, G. H., Kowalski, L., and McGowan, R., Lack of predictability in rats to exhibit chronic oxygen poisoning. Aerospace Med. 36, 634-366 (1964). K14. Kylstra, J. A., Breathing fluid. Experientia 18, 1-2 (1962). K15. Kylstra, J. A., Drowning: The role of salts in the drowning fluid. A preliminary report on observations in mice. Acta Physwl. Pharmacol. Neerl. 10, 327-334 (1962). K16. Kylstra, J. A., Hyperbaric oxygenation of small mammals. Lancet 11, 149 (1962). K17. Kylstra, J. A., Tissing, M. O., and Van der Maen, A., Of mice as fish. Trans. Am. SOC.ArtiJicial Internal Organs 8, 378-379 (1962). K18. Kylstra, J. A., Survival in air after breathing fluid. Lancet 11, 1170 (1962). K19. Kylstra, J. A., and Lanphier, E. H., Gas exchange in fluid ventilated dogs. Federation Proc. 23, 469 (1964). Ll, Lambertsen, C. J., Kough, R. H., Cooper, D. J., Emmel, G. L., Loeschke, H. H., and Schmidt, C. F., Oxygen toxicity. Effects in man of oxygen inhalation a t 1 and 3.5 atmospheres upon blood gas transport, cerebral circulation, and cerebral metabolism. J . Appl. Physwl. 6, 471-486 (1953). L2. Lambertsen, C. J., Stroud, M. W., Gould, R. A., Kough, R. H., Ewing, J. H., and Schmidt, C. F., Oxygen toxicity. Respiratory responses of normal men t o inhalation of 6 and 100% oxygen under 3.5 atmospheres pressure. J . Appl. Physwl. 6, 487-494 (1953).
132
BHELDON F. GOWLIEB
L3. Lambertsen, C. J., Ewing, J. H., Kough, R. H., Gould, R. A., and Stroud, M., 111. Oxygen toxicity. Arterial and internal jugular blood gas composition in man during inhalation of air, 1 0 0 ~ O2 o and 2’3, COz in 02 a t 3.5 atmospheres ambient pressure. J . Appl. Physiol. 8, 255-263 (1955). L4. Lanphier, E. H., Diving medicine. Neu, Engl. J . Med. 266, 120-131 (1957). L5. Lanphier, E. H., and Morin, R. A,, Effects of gas density on carbon dioxide elimination. Physiologist 4, 63 (1961). L6. Lanphier, E. H., and Busby, D. E., Alveolar and arterial Peon in man under increased ambient pressures. Proc. 99nd Intern. Congr. Intern. Unwn Physwl. Sci., Leiden, 1969 Vol. 2, p. 301 (abstract). Excerpta Med. Found. Amsterdam, 1962. L7. Lawrence, J. H., Loomis, W. F., Tobias, C. A., and Turpin, F. H., Preliminary observations on the narcotic effect of xenon with a review of values for solubilities of gases in water and oils. J. Physiol. (London) 106, 197-204 (1946). L8. Lawson, D. D., McAllister, R. A., and Smith, G., The effect of high pressure oxygen in experimental acute carbon monoxide poisoning. Scot. Med. J . 4, 327 (1959). L9. Lawson, D. D., McAllister, R. A., and Smith, G., Treatment of acute experimental carbon-monoxide poisoning with oxygen under pressure. Lancet I, 800-802 (1961). L10. Lazerev, N. V., Lyublina, E. T., and Madorskaya, R. Y., Narcotic action of xenon. Fizol. Zh. SSSR 84, 131-134 (1948). L11. Leather, R. P., and Eckert, C., Hyperbaric oxygenation and mechlorethamine effectiveness. Arch. Surg. 87, 144-147 (1963). L12. Leon, H. A., and Cook, S. F., A mechanism by which helium increases metabolism in small animals. Am. J . Physiol. 199, 243-245 (1960). L13. Levine, S., Oxygen in the therapy of cyanide poisoning. J . Am. Med. Assoc. 170, 1585 (1959). L14. Levy, F., and Featherstone, R. M., The effect of xenon and nitrous oxide on in vitro guinea pig brain respiration and oxidative phosphorylation. J. Pharmacol. Exptl. Therap. 110, 222-225 (1954). L15. Linderstrem-Lang, K., Principles of the cartesian diver applied to gasometric technique. Nature 140, 108 (1937). L16. Lippincott, C. L., and Harter, W. L., Oxygen saturation under pressure as a treatment for tetanus in a dog. J . Am. Vet. Med. Assoc. 142, 872-874 (1964). M1. Mallams, J. T., Finney, J. W., and Balla, G. A., The use of hydrogen peroxide as a source of oxygen in a regional intra-arterial infusion system. Southern Med. J . 66, 230-232 (1962). M2. Manax, W. G., Block, J. H., Longerbeam, J. K., and Lillehei, R. C., Successful 24 hour in vitro preservation of canine kidneys by the combined use of hyperbaric oxygenation and hypothermia. Surgery 66, 275-282 (1964). M3. Marshall, W. H., Hoppe, E. T., and Stark, F., The effect of ambient oxygen tension on the toxicity and therapeutic effect of mechlorethamine (nitrogen mustard). Arch. Surg. 86, 932-939 (1963). M4. Marsland, D. A., and Brown, D. E., Tae effects of pressure on sol-gel equilibrium with special reference to myosin and other protoplasmic gels. J . Cellular Comp. Physwl. 20, 295-305 (1942). M5. Matteo, R. S., and Nahas, G. G., Sodium bicarbonate: increase in survival rate of rats inhaling o-tygen. Science 141, 719-720 (1963). M6. McAllister, R. A., %ark, J. M., Norman, J. N., and ROES,R. M., Inhibitory effecta of hyperbaric oxygen on bacteria and fungi. Lancet 11, 1040-1042 (1963).
HYPERBARIC OXYGENATION
133
M7. McDowall, D. G., Anaesthesia in a pressure chamber. Aaaesthesia 19, 321-336 (1964). M8. “Medical Electronic News,’’ Vol. 4, Nos. 1, 2, 3. Instr. Publ. Co., Pittsburgh, Pennsylvania, 1964. M9. Meijne, N. G., Bulterijs, A. B., Schoemaker, G., and Eloff, S. J. P., Treatment of dogs with oxygen under high pressure after ligation of the descending branch of the left coronary artery. Diseases Chest 44, 234-250 (1963). M10. Meijne, N. G., Bulterijs, A. B., Eloff, S. J. P., and Boerema, I., An experimental investigation into the influence of administration of oxygen under increased atmospheric pressure upon coronary infarction. J. Cardiovascular Surg. 4, 521-535 (1963). M11. Meijne, N. G., The influence of high nitrogen tensions on workers in compressed air. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 323-329. Elsevier, Amsterdam, 1964. M12. Meijne, N. G., The safe period of circulatory arrest at 3 ata. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and Meijne, N. G., eds.), pp. 201-204. Elsevier Amsterdam, 1964. M13. Meijne, N. G., Schoemaker, G. and Bulterijs, A. B., Oxyg’en supply t o ischaemic myocardial tissue under increased atmospheric pressure. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne eds.), pp. 69-74. Elsevier, Amsterdam, 1964. M14. Meijne, N. G., Schoemaker, G., and Bulterijs, A. B., The value of hyperbaric oxygen in cardiovascular surgery. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 182-188. Elsevier, Amsterdam, 1964. M15. Mengel, C. E., Kann, H. E., Smith, W. W., and Horton, B. D., Effects of in Viva hyperoxia on erythrocytes. 1.Hemolysis in mice exposed to hyperbaric oxygenation. Proc. SOC.Exptl. Bwl. Med. 116, 259-261 (1964). M16. Michel, E. L., Langevin, R. W., and Gell, C. F., Effect of continuous exposure to oxygen tension of 418 mm Hg for 168 hours. Aerospace Med. 31, 138-144 (1960). M17. Moore, B., and Williams, R. S., The growth of B . tubercubsia and other microorganisms in different percentages of oxygen. Biochem. J. 4, 177-190 (1908). M18. Moore, B., and Williams, R. S., The growth of various species of bacteria and other microorganisms in atmospheres enriched with oxygen. Biochem. J. 6 , 181-187 (1911). M19. Morita, R. Y., and Height, R. D., Malic dehydrogenase activity a t 101°C under hydrostatic pressure. J. Bacterial. 83, 1341-1346 (1962). M20. Moss, P. D., High pressure oxygen in the treatment of neonatal anoxia. Arch. Disease Childhood 37, 452 (1962). M21. Moulton, G. A., Esmond, W. G., and Michaels, M., Effect of hyperbaric oxygenation on Noble-Collip drum shock in the rat. Bull. School Med. Univ. Maryland 47, 42-44 (1962). M22. Muller, D. A., Clostridium.welchii infection in gynaecology. S. African Med. J . s8, 539-541 (1964). M23. Mullins, L. J., Ion transfer in cells under high gas tensions. Exptl. Cell Res. Suppl. 1, 328-330 (1949). N1. National Research Council., In “Manual of DecompressionSickness” (J.F. Fulton, ed.), p. 437. Saunders, Philadelphia, Pennsylvania, 1951.
134
SHELDON F. OOTTLIEB
N2. Nelson, C. F., and Woodward, P., The relief of experimental anoxemia by compressed air. J. Pathol. Bacteriol. 28, 507-513 (1925). N3. Nilsson, N. J., Oximetry. Physiol. Rev. 40, 1-26 (1960). N4. Nuckolls, J. G., and Osterhout, S., The effect of hyperbaric oxygen on anaerobic bacteria. Clin. Res. 12, 244 (1964). 01. Ohlssen, W. T. L., A study on oxygen toxicity a t atmospheric pressure with special references to the pathogenesis of pulmonary damage and clinical oxygen therapy. Acta Med. Scand. Suppl. 190, 1-93 (1947). P1. Pace, N., Strajman, E., and Walke, E. L., Acceleration of carbon monoxide elimination in man by high pressure oxygen. Science 8, 652-654 (1950). P2. Pacheco, G., and Costa, G. A., Effect of oxygen under pressure on Clostridium welchii. Rev. Brasil. Biol. 1, 145-153 (1941). P3. Pascale, L. R., Wallyn, R. J., Goldfein, S., and Gumbinar, S. H., Observations in response of tetanus to hyperbaric oxygenation. I n “Clinical Application of Hyperbaric’Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 44-51. Elsevier, Amsterdam, 1964. P4. Pascale, L. R., Wallyn, R. J., Goldfein, S., and Gumbinar, S. H., Treatment of tetanus by hyperbaric oxygenation. J. Am. Med. Assoc. 189,408410 (1964). P5. Paul, J., and Danes, B. S., A modified Cartesian diver method permitting measurement of oxygen uptake in the presence of carbon dioxide. A n d . Chem. 82, 470-485 (1961). P6. Paulet, G., Valeur et mechanisme d’action de l’oxygbneth6rapie dans le traitement de l’intoxication cyanhydrique. Arch. Intern. Physwl. Bwchim. 63, 340-360 (1955). P7. Pauling, L., Wood, R. E., and Sturdivant, J. H., An instrument for determining the partial pressure of oxygen in a gas mixture. Science 108, 338 (1946). P8. Pegg, J., Homer, T., and Wahrenbrock, E., Mammalian respiration of pressureoxygenated solutions. Physiologist 6, 194 (1962). P9. Penrod, K. E., EfEect of intermittent nitrogen exposures on tolerance to oxygen a t high pressures. A m . J. Physiol. 186, 149-151 (1956). P10. Perot, P. L., Jr., and Stein, S. N., Conduction blockade in peripheral nerve produced by oxygen a t high pressure. Science 128, 802-803 (1956). P11. Peters, J. P., and Van slyke, D. D., “Quantitative Clinical Chemistry” Williams & Wilkins, Baltimore, Maryland, 1932. P12. Petropoulos, P., Influence of hyperbaric oxygenation on the haemodynamio changes and mortality after circumflex coronary artery occlusion. I n “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 84-99. Elsevier, Amsterdam, 1964. P13. Pfeiffer, C. C., and Gersh, I., The prevention of the convulsions of oxygen poisoning by means of drugs. Naval Med. Res. Inst. (Bethesda) Rept. No. 2, 1-8 (1944). P14. Pittinger, C. B., Faulkener, A., Knott, J. R., Pinder, J., Morris, E., and Bickford, R. G., Electroencephalographic and other observations in monkeys during xenon anesthesia a t elevated pressures. Anesthesiology 16, 551-563 (1955). P15. Pittinger, C. B., Considerations of hyperbaric inhalation anesthesia. Southern Med. J. 67, 395-398 (1964). P16. Pitts, G. C., and Pace, N., The effects of breathing oxygen at 2.5 atmospheres on the rate of elimination of carbon monoxide in man. Naval Med. Res. Inst. (Bethesda) Proj. No. NM001-056-01.4 (1949). P17. Popovic, V., Gerschman, R., and Gilbert, D. L., Effect of high oxygen pressure on
HYPERBARIC OXYGENATION
135
ground squirrels in hypothermia and hibernation. Am. J . Physwl. 208, 49-60 (1964). P18. Pratt, P., Pulmonary capillary proliferation induced by oxygen inhalation. Am. J . Pathol. 34, 1033-1050 (1958). P19. Pritchard, G. G., The effect of high oxygen pressure on the respiratory metabolism of pea seeds. J . Ezptl. Botany 12, 353-363 (1961). R1. Rahn, H., and MacHattie, L., Survival of mice in absence of inert gas. Proc. SOC. EzpU. B i d . Med. 104, 772-775 (1960). R2. Richards, V., Pinto, D., and Coombs, P., Considerations and uses of hyperbaric oxygen therapy in surgery. Am. J . Surg. 106, 114-127 (1963). R3. Roahe, J., and Allen, W., Effects of hyperbaric oxygenation on left circumflex coronary artery occlusion in dogs. Surg. Forum 16, 208-209 (1964). R4. Roth, E. M., Kann, H. E., Jr., Mengel, C. E., Smith, W., and Horton, B., Oxygen toxicity and vitamin E. Aerospace Med. 36, 84&844 (1964). S1. Sanger, C., Churchill-Davidson, I., and Thomlimon, R. H., Anesthesia for radiotherapy under high pressure oxygen. Brit. J . Anaesthesia 27, 436-446 (1955). 52. Sanger, C., High pressure oxygen and radiation therapy. Am. J . Roentgenol. Radium Therapy Nucl. Med. 81, 498-503 (1959). 53. Sanger, C., Nahas, G. G., Goldberg, A. R., and D’Alessio, G. M., Effects of 2-amino2-hydroxymethyl-1,3-propanediol on oxygen toxicity in mice. Ann. N . Y. Acad. Sci. 92, 71CL723 (1961). 84. Schaffner, F., Lee, U. L., Jr., and Schildkraut, H. S., Hepatic changes after breathing pure oxygen. Federation Proc. 23, 522 (1964). 55. Schnedorf, J. G., and Orr, T. G., Beneficial effects of oxygen therapy in experimental traumatic shock. Surg. Bynecol. Obstet. 73, 79-83 (1941). S6. Schnedorf, J. G., and Orr, T. G., Oxygen therapy in shock due to hemorrhage. Surg. Gynecol. Obstet. 73, 495-497 (1941). 57. Schoemaker, G., Oxygen tension measurements under hyperbaric conditions. In “Clinical Applications of Hyperbaric Oxygen” (I.Boerema, W. H. Bnunmelkamp, and N. G. Meijne, eds.), pp. 330-335. Elsevier, Amsterdam, 1964. SS. Schultz, H., Changes in shape of the mitochondria in the alveolar epithelium during respiration in carbon dioxide (-air mixture) and oxygen. Naturwiss. 43, 205-206 (1956). S9. Schwartz, S. I., and Brwlau, R. C., Protective effect of hyperbaric oxygenation during thoracic aortic occlusion. I n “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 395-402. Elsevier, Amsterdam, 1964. S10. Sergeant, R. L., Speech during respiration of a mixture of helium and oxygen. Aerospace Med. 34, 826-829 (1963). 811. Severinghaus, J. W., and Bradley, A. F., Electrodes for blood p0.r and pC0, determination. J . Appl. Physwl. 13, 515-520 (1958). 512. Severinghaus, J. W., Electrodes for blood and gas pC02, pO2 and blood pH. Acta Aaaestheswl. Scand. 11, 207-220 (1962). S13. Sevcik, V., and Wymola, F., The effect of oxygen on the course of model anaerobic infection. J. Hyg. Epidemiol. Microbial. Immunol. (Prague) 8, 313-317 (1964). 514. Sluyter, M. E., Carbon monoxide poisoning. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.) pp. 166-167. Elsevier, Amsterdam, 1964.
136
SHELDON F. GOTTLIEB
515. Smith, C. W., Bean, J. W., and Bauer, R., Thyroid influence in reactions to oxygen a t atmospheric pressure. Am. J . Physwl. 198, 883-888 (1960). S16. Smith, F. I. C., Heim, J. W., Thomson, R. M., and Drinker, C. K., Bodily changes and development of pulmonary resistance in rats living under compressed air conditions. J . Exptl.Med. 66, 63-78 (1932). 517. Smith, G., and Lawson, D. D., Experimental coronary occlusion effects of the administration of oxygen under pressure. Scot. Med. J. 3, 346-350 (1958). 518. Smith, G., and Sharp, G. R., Treatment of carbon monoxide poisoning with oxygen under pressure. Lancet 11, 905-906 (1960). S19. Smith, G., Lawson, D. D., Redrew, S., Ledingham, I. McA., and Sharp, G. R., Preservation of cerebral cortical activity by breathing oxygen a t two atmospheres of pressure during cerebral ischemia. Surg. Gynecol. Obstet. 113, 13-16 (1961). 520. Smith, G., Stevens, J., Graths, J. C., and Ledingham, I. McA., Near-avulsion of foot treated by replacement and subsequent prolonged exposure of patients to oxygen a t two atmospheres pressure. Lancet 11, 1122-1123 (1961). 521. Smith, G., and Lawson, D. D., “he protective effect of inhalation of oxygen a t two atmospheres absolute pressure in acute coronary arterial occlusion. Surg. Gynecot. Obstet. 144, 320-322 (1962). S22. Smith, G., Ledingham, I. McA., Sharp, G. R., Norman, J. N., and Bates, E. H., Treatment of coal-gas poisoning with oxygen a t 2 atmospheres pressure. Luncet I, 816-818 (1962). 523. Smith, G., Sillar, W., Norman, J. N., Ledingham, I. McA., Bates, E. H., and Scott, A. C., Inhalation of oxygen a t 2 atmospheres for Clostridium welchii infections. Lancet 11, 75G757 (1962). 524. Smith, J. L., The pathological effects due to increase of oxygen tension in the air breathed. J. Physwl. (London) 24, 19-35 (1899). S25. Smith, R. M., Crocker, D., and Adams, J. C., A’nesthetic management of patients during surgery under hyperbaric oxygenation. Anesthesia Analgesia Current Res. 43, 766-776 (1964). S26. Stadie, W. C., Riggs, B. C., and Haugaard, N., Oxygen poisoning. A m . J . Med. Sci. 207, 84-114 (1944). S 7 . Stadie, W. C., and Haugaard, N., Oxygen poisoning. V. The effect of oxygen pressure upon enzymes: succinic dehydrogenase and cytochrome oxidase. J . Bwl.Chem. 161, 153-174 (1945). S28. Stadie, W. C., Riggs, B. C., and Haugaard, N., Oxygen poisoning. 111. The effect of high oxygen pressures upon the metabolism of brain. J . Bwl.Chem. 160, 191-208 (1945). 529. Stadie, W. C., Riggs, B. C., and Haugaard, N., Oxygen poisoning. IV. The effect of high oxygen pressure upon the metabolism of liver, kidney, lung and muscle tissue. J . Biol. Chem. 160, 209-216 (1946). S30. Stadie, W. C., Riggs, B. C., and Haugaard, N., Oxygen poisoning. VI. The effect of high oxygen premure upon enzymes: pepsin, catalase, cholinesterase, and carbonic anhydrase. J . Biol. Chem. 161, 175-180 (1945). 531. Stadie, W. C., and Haugaard, N., Oxygen poisoning. VII. The effect of high oxygen pressure upon enzymes :uricase, xanthine oxidase, and D-amino acid oxidase. J. Biol. Chem. 161, 181-188 (1945). S32. Stadie, W. C., Riggs, B. C., and Haugaard, N., Oxygen poisoning. VIII. The effect of high oxygen pressure on enzymes: the system synthesizing acetylcholine. J. Biol. Chem. 161, 189-196 (1945).
HYPERBARIC OXYGENATION
137
S33. Stow, R. W., Baer, R. F., and Randall, B. F., Rapid measurement of the tension of carbon dioxide in blood. Arch. Phys. Med. Rehabil. 38, 64&650 (1957). S34. Stuart, B., Gerschman, R., and Stannard, J. N., Effect of high oxygen tension on potassium retentivity and colony formation of baker’s yeast. J. Gen. Physwl. 46, 1019-1030 (1962). S35. South, F. E., Jr., and Cook, S. F., Effect of helium on the respiration and glycolysis of mouse liver slices. J . Gen. Physwl. 36, 513-528 (1953). S36. South, F. E., Jr., and Cook, S. F., Argon, xenon, hydrogen and the oxygen consumption and glycolysis of mouse tissue slices. J . Gen. Physio2. 37, 335-341 (1954). T1. Tanaka, R., Fujimori, M., and Virtue, R. W., Oxygen utilizat,ion by dogs after administration of potassium perchlorate during hypothermia and a t a pressure of two atmospheres. Anesthesiology 22, 20-23 (1961). T2. Thaysen, A. C., Action of oxygen under pressure at various temperatures. Biochem. J. 28, 1330-1335 (1934). T3. Thomas, J. J., Jr., Neptune, E. M., Jr., and Suddath, H. C., Toxic effects of oxygen at high pressure on the metabolism of D-glucose by dispersions of rat brain. Bhchem. J. 88, 31-44 (1963). T4. Thomlinson, R. H., An experimental met.hod for comparing treatment of intact malignant tumors in animals and its application to the use of oxygen in radiotherapy. Brit. J. Cancer 14, 555-576 (1960). T5. Trowell, 0. A., The effect of environmental factors on the radiosensitivity of lymph nodes cultured in vitro. Brit. J . Radwl. 26, 302-309 (1953). T6. Turner, E. R., and Quartley, C. E., Studies in the respiratory carbohydrate metabolism of plant tissues. VIII. An inhibition of respiration in peas induced by “oxygen poisoning.” J. Exptl. Botany 7 , 362-371 (1956). U1. Umbreit, W. W., Burris, R. H., and Stauffer, J. F., “Manometric Techniques and Tissue Metabolism.” Burgess, Minneapolis, Minnesota, 1951. U2. U.S. Navy, I. General principles of diving, navyships, In “U.S. Navy Diving Manual,” pp. 250-538. Department of the Navy,Washington, D.C., 1959. V1. Van den Brenk, H. A. S., Effect of high pressure oxygen on radiosensitivity of Ehrlich’s tumor in mice after “immunological approximation.” Brit. J. Cancer 16, 61-84 (1961). V2. Van den Brenk, H. A. S., Elliot, K., and Hutchings, H., Effect of single fractionated doses of x-rays on radiocurability of solid Ehrlich tumour and tissue reactions in vivo, for different oxygen tensions. Brit. J. Cancer 16, 518-534 (1962). V3. Van den Brenk, H. A. S., and Jamieson, D., Pulmonary damage due to high pressure oxygen breathing in rats. 1, Lung weight, histological and radiological studies. Australian J . Exptl. Biol. Med. Sci. 40, 37-49 (1962). V4. Van den Brenk, H. A. S., and Jamieson, D., Potentiation by anesthetics of brain damage due to breathing high-pressure oxygen in mammals. Nature 194, 777-778 (1962). V5. Van den Brenk, H. A. S., Elliot, K., and Hutchings, H., Further observations on radiocurability of a solid Ehrlich tumour and tissue reactions in the mouse with fractionated radiation doses and effects of oxygen. Brit. J. Cancer 17, 281-288 (1963). V6. Van den Brenk, H. A. S., and Jamieson, D., Brain damage and paralysis in animals exposed to high pressure oxygen-pharmacological and biochemical observations. Biochem. Pharmacol. 13, 165-182 (1964). V7. Van den Brenk, H. A. S., Madigan, J. P., and Kerr, R. C., Experience with mega-
138
SHELDON F. GOTTLIEB
voltage irradiation of advanced malignant disease using high pressure oxygen. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H., Brummelkamp, and N. G. Meijne, eds.), pp. 144-160. Elsevier, Amsterdam, 1964. V8. Van Elk, J., and Trippel, 0. H., Ventricular fibrillation following coronary artery occlusion in the dog heart and the possible protective effect of high pressure oxygen. I n “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 116-120. Elsevier, Amsterdam, 1964. V9. Van Slyke, D. D., and Neill, J. M., The determination of gases in blood and other solutions by vacuum extraction and manometric measurement. I. J. Biol. Chem. 61, 523-573 (1924). VlO. Vermeulen-Cranch, D. M. E., Anesthesia in a high pressure chamber. In “Clinical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.) pp. 205-208. Elsevier, Amsterdam, 1964. V11. Volskii, M. I., The assimilation of nitrogen by animal organisms as exemplified by chicken embryos and honeybee pupae. Dok1.-Biol. Sci. Sect. (English Trawl.) 128, 895-897 (1960). Wl. Walker, I. G., The involvement of carbon dioxide in the toxicity of oxygen at high pressure. Can. J. Biochem. Physiol. 38, 1803-1810 (1961). W2. Wallyn, R. J., Gumbinar, S. H., Goldfein, S., and Pascale, L. R., The treatment of anaerobic infections with hyperbaric oxygen, Surg. C1i:a. North Am. 44, 107-112 (1964). W3. Weaie, F. E., The place of oxygen in shock. Brit. J . Anaesthesia 32, 76-80 (1960). W4. Weale, F. E., Pkessurized oxygen for total body perfusion, Lancet 11, 570-573 (1961). W5. Weir, F. W., Bath, D. W., Yevich, P., and Oberst, F. W., A study of the effects of continuous inhalation of high concentrations of oxygen at ambient pressure and temperature. U.S. Army Chem. Res. Develop. Lab. (Army Chem. Center, Maryland), Proj. No. 7165: task 716401, 1-12 (1961). W6. West, J. B., Fowler, K. T., Hugh-Jones, P., and O’Donnell, T. V., Measurement of the ventilation-perfusion ratio inequality in the lung by the analysis of a single expirate. Clin. Sn‘. 16, 529-549 (1957). W7. West, J. B., and Hugh-Jones, P., Patterns of gas flow in the upper bronchial tree. J. Appl. Physiol. 14, 743-759 (1959). W8. West, J. B., Measurement of bronchial air flow. J. Appl. Physiol. 16, 976-978 (1960). W9. White, J. H., Report of case of gas gangrene treated with oxygen injection and peroxide. J. Oklahoma State Med. Assoc. 21, 59-60 (1928). W10. Wildemuth, O., Hybaroxic radiation therapy in cancer management. Radiology 82, 767-777 (1964). W11. Williams, J. W., The difference in growth of pathogenic fungi with variation of medium and oxygen tension. J . Lab. Clin. Med. 24, 39-43 (1938). “12. Winkel, C. A., and Kroon, T. A. J., Experience with hyperbaric oxygen treatment in tetanus. Ita “&nical Application of Hyperbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 52-62. Elsevier, Amsterdam, 1964. W13. Wolman, M., and Zaidel, L., Hyperoxia and formation of chromolipoid pigments. Ezperientia 18, 1-3 (1962). W14. Wood, G. O., Mason, M. F., and Blalock, A., Studies on effects of high concentrations of oxygen in experimental shock. Surgery 8, 247-256 (1940). W15. Wood, J. D., and Watson, W. J., Gamma-aminobutyric acid levels in the brain
HYPERBARIC OXYGENATION
139
of rats exposed to oxygen at high pressures. Can. J. Bwchem. Physiol. 41, 1907-1913 (1963). W16. Wood, J. D., and Watson, W. J., Molecular structure-activity relationships of compounds protecting rats against oxygen poisoning. Can. J . Physwl. Pharmacol. 43, 641-646 (1964). W17. Wright, G. P., Neurotoxins of Clostridium botulinum and Clostridium tetani. Pharmacol. Rev. 7 , 413465 (1955). Y1. Yanda, R. L., Motley, H. L., and Smart, R. H., The effects of pressure upon lung volumes of pulmonary emphysema patients and upon normal individuals. In “Clinical Application of Hyerbaric Oxygen” (I. Boerema, W. H. Brummelkamp, and N. G. Meijne, eds.), pp. 336-345. Elsevier, Amsterdam, 1964. Y2. Yarbrough, 0. D., Welham, W., Brinton, E. S., and Behnke, A. R., Symptoms of oxygen poisoning and limits of tolerance a t rest and work. U.S. Naval Ezptl. Diving Unit (Washington, D.C.), Proj. No. X-337, Sub. No. 62, Rept., 1 (1947).
This Page Intentionally Left Blank
DETERMINATION OF HEMOGLOBIN AND ITS DERIVATIVES E. J. van Kampen a n d W. G. Zijlstra With the collaboration of 0. W. van Assendelft and W. A. Reinkmgh. Clinical Chemical Laboratory, Diaconessenhuis, and laboratory of Chemical Physiology, University of Groningen, The Netherlands
1. Introduction.
.
...................
1.1. General Requirements for Hb Determination and Standardtation.. . . . 1.2. General Aspects of the Determination of Hemoglobin Derivatives. . . . . 2.2. The Hemiglobincyanide Method. 2.3. Preparation of a Hemiglobincyan 3. Absorption Spectra and Extinction Coe
4. 5.
6.
7.
143 145
. . . . . . . . . . . . . . . 154
3.2. Preparation of Blood Samples. ...... ..................... 3.3. The Absorption Spectrum of Hemoglobin (Hb). ..................... 3.4. The Absorption Spectrum of Oxyhemoglobin (HbOs).. . . . . . . . . . . . . . . . 3.5. The Absorption Spectrum of Carboxyhemoglobin (HbCO) . . . . . . . . . . . . 3.6. The Absorption Spectrum of Hemiglobin (Hi). ...................... 3.7. The Absorption Spectrum of Hemiglobincyanide (HiCN). . 3.8. Light Absorption by Non-Hemoglobin.. ............ General Aspects of Two-Wavelength Methods. ........... Spectrophotometric Determination of the Oxygen Saturation. .............. 5.1. Determination of SO, Meaaured a t X = 660 and 522 nm, and X = 560 and 506 nm,Respectively; 2 = 0.01 cm.. .......................... 5.2. Comparison of the Two Methods.. ................................ 5.3. Comparison of SO, Measured Spectrophotometrically at X = 560 and 522 nm and Determined by Van Slyke’s Manometric Method.. . . . . . . . Spectrophotometric Determination of Carboxyhemoglobm. ................ 6.1. Determination of So0 in a Mixture Containing Hb and HbCO Measured ......... a t X = 535 and 578 nm; 1 = 1.000 cm 6.2. Determination of SCOin a Mixture Containing HbOz and HbCO Measured a t A = 562 and 540 nm; I = 0.01 cm.. . . . . . . . . . . . . . . . . . . . . . . . . ....... 6.3. Comparison of the Two Methods.. . . . . . . . . . . . . . . . . . Spectrophotometric Determination of Hemiglobin. . . . . . . . . . ......................... 7.1. Principles.. ......................... 7.2. Determination of Measured a t X = and 523 nm; 1 = 0.01 cm.. . 7.3. Determination of SEi by Converting Hi to HiCN and Measuring a t X = 630nm ..................................................... 7.4. Estimation of Sai by a Screening Method.. . . . . . . . . . . . . . . . . . . . . . . . . . 7.5. Comparison of the Three Methods. ............................ 141
156 156 158 160 162
167 168 169 170 170 172 174 176 177 178 178 179 181
142
E. J. VAN KAMPEN AND W. G. ZIJLSTRA
8. Spectrophotometric Determination of Sulfhemoglobin..................... 8.1. Estimation of &Hb by a Screening Method.. ........................ References. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
182 184 185
1. Introduction
The measurement of the total hemoglobin concentration and the determination of the hemoglobin derivatives oxyhemoglobin (HbOz), carboxyhemoglobin (HbCO), sulfhemoglobin (SHb), and hemiglobin (Hi) in blood are undoubtedly the most frequent determinations in the clinical chemical laboratory. In contrast to nearly all other determinations, a clinical diagnosis is often made solely on the basis of hemoglobin determinations: anemia when values are too low, polycythemia in cases of abnormally high values.
I
I
I
I
I
I
I
I
I
1
2
3
4
5
6
7
8
9 t(months)
FIG. 1. Relationship between mean monthly hemoglobin concentration (CHL)and number of pints of blood (P) transfused in a general hospital. Arrow indicates standardization of hemoglobin determination. Based on data from Mann et a2. (M3).
When significant variations occur in the determination of the hemoglobin concentration, it becomes difficult to give a clear-cut definition of, for example, anemia. The effects of a change in the measurement technique of hemoglobin concentration is clearly illustrated in Fig. 1, where the number of pints of blood transfused in a certain general hospital fell drastically as soon as the hemoglobin determination was standardized. This one example alone indicates that the standardization of hemoglobinometry is absolutely necessary.
HEMOGLOBIN AND ITS DERIVATIVES
143
FOR HB DETERMINATION STANDARDIZATION
1 .l. GENERALREQUIREMENTS AND
1.1.1. Method of Choice
As hemoglobin is itself a colored substance and because a spectrophotometric technique is generally the method of choice in the clinical chemical laboratory, it is logical t o prefer a photometric technique for the determination of the Hb concentration. The first general requirement therefore must be that Lambert-Beer's law is applicable to the method: log (100/T) = D = ECZ (1)
where T = light transmission in %, D = optical density, e = millimolar extinction coefficient, c = concentration in millimoles per liter, and 1 = layer thickness in cm. 1.1.2. Dilution Factor
A second general requirement must be that the dilution factor be so chosen as to ensure an acceptable ratio between blood and reagent. This ratio (e.g., 0.01-0.05 ml of blood, 5.0 ml of reagent) must be acceptable to the clinical chemist while at the same time the extinction must be optimal from the physicochemical view. This requires that the extinction should be found to lie in that region where the random error (dclc) is smallest. It follows from Eq. (1) that log (100/T) c= (2) €1 and after transposing to natural logarithms In 100 - In T In 10
= EC1
Differentiating T with respect to c gives
( 1 / T )dT
=
--a!
dc or dc
=
- (l/aT) dT
where a! = e I . In 10. Dividing by Eq. (2), after transposing to natural logarithms, gives
dc= C
1 T In 100/T dT
Now ( d c / d T ) / c is minimal when
dT "[-
T In 100/T ]
=o
144 or
E. J. VAN EAMPEN AN'D W. 0. ZIJLSTRA
+
In 100 - ( T ( l / T ) In 7') ( T In 100/T)2
=o
thus In 100 - 1 - In T = 0 In T = 3.6 T = 36.6% The most exact spectrophotometric measurements will thus be performed around T = 36.6%, which equals l/e, or D = 0.434. Figure 2 presents the
IDOO aeooa600 woo a300 0.200 0100 0.050 O ~ ~ - ~ - ~ - - ~ - - - - ~ - - - I - - - - ~ - - - - ~ - - ~ - - - - - - -
D
Corresponding optical densities are also indicated ( D ) . Graph based on Eq. (31, taking dT = 1%.
relationship between dc/c (random error in yo)and T (yo).From this it is clear that the dilution factor must be so chosen as to ensure measurements in the region of 20% < T < 80% or 0.090 < D < 0.700.
1.1.3. Absorption Spectrum The light absorption spectrum of the hemoglobin derivative to be chosen for a photometric determination should have a number of favorable properties. It should display a rather flat maximum. If this is not the case, slight alterations in filter or in photometer characteristics will introduce considerable errors. Furthermore, the light absorption maximum should
HEMOGLOBIN AND ITS DERIVATIVES
145
lie in the spectral region 400 nm < X < 600 nm and be independent of changes in the pH. Hemiglobincyanide seems to meet perfectly the requirements stated above. The light absorption spectrum of hemiglobincyanide is given in Fig. 3. D
-
2.000 1.000
0.800 0.600 0.500 0.100
0.300 0.200 0.100
0
680
640
600
560
520
180 h(nrn)
FIQ.3. D/X curve of hemiglobincyanide (HiCN), prepared as described in Section 2.3. Recorded on a Beckman DB spectrophotometer, layer thickness 1.000 cm. 1.1.4.
Specijicity and Stability
Because measurements dependent upon time always disturb laboratory routine, the hemoglobin derivative to be used for the determination of the Hb concentration in blood must be formed rapidly, all hemoglobins present in the sample must be converted, and the end product formed must be stable. Stability is an essential requisite for the use of standard solutions to calibrate the photometer to be used. The measurement itself should not be disturbed by plasma proteins (globulins) or erythrocyte stromata, either through the introduction of a so-called protein error (C3) or because of turbidity. Only HiCN properly prepared fulfills all requirements mentioned.
GENERALASPECTSOF THE DETERMINATION OF HEMOGLOBIN DERIVATIVES The introduction of modern spectrophotometers, which allow measurements with a small half-intensity band width (1 nm)’, raised the possibility of applying photometric methods t o systems of two or more components. The primary conditions for this are (a) a distinct difference between the 1.2.
1 nm = 1 nanometer = 10” meter.
146
E. J. VAN KAMPEN AND W. Q. ZIJLSTRA
absorption spectra of the components, and (b) the applicability of LambertBeer’s law to each of the components. If a solution contains n dissolved substances, complete determination of this n-component system requires measurements at at least n different wavelengths. These measurements provide n equations of the type
D A= E
+ ECCJ+ . . + e n k l
~ c I ~
(4)
where Dh= optical density (log l o / ] a)t wavelength A, enh = extinction coefficient of component n at wavelength A, c,, = concentration of component n, and 1 = layer thickness. The values of c1, c2, . . . , cn can be calculated by solving these equations, if the extinction coefficients (n2in number) are known. Extinction coefficients of a number of hemoglobin derivatives are presented in Section 3. Although these values could be used in the actual analysis of n-component systems, it seems as yet unwise to discontinue using the simple two-wavelength methods which have long been in use with reasonable results. Thus, these established methods are described in Sections 5, 6, and 7, although it now seems probable that future developments will tend to the general use of methods involving the actual extinction coefficients. 2. The Standardized Determination of Hemoglobin
2.1. THEEXTINCTION COEFFICIENT OF HEMIGLOBINCYANIDE AT A = 540 NM; e&& 2.1.1. Introduction
The fundamental quantity to be used in any standardized spectrophotometric determination of hemoglobin should be the quarter millimolar extinction coefficient because of the four Fe atoms per hemoglobin molecule. This quantity is independent of the molecular weight of hemoglobin. Consequently, a redetermination of the molecular weight will not change the value of the extinction coefficient to be used and will not affect hemoglobin concentrations when these are expressed in quarter millimoles per liter. When the value of the quarter millimolar extinction coefficient of hemiglobincyanide a t A = 540 nm (EF&)is determined using hemoglobin solutions prepared from washed red cells by toluene hemolysis, the solutions may possibly contain the following substances: (1) hemoglobin (Hb), oxyhemoglobin (HbOz),carboxyhemoglobin (HbCO), hemiglobin (Hi), and other hemoglobin derivatives (“Hx”), all convertible to HiCN; (2) hemo-
HEMOGLOBIN AND ITS DERIVATIVES
147
globin degeneration products, other compounds (some of which contain iron), and traces of free iron (non-HiCN). Spectrophotometry of a hemoglobin solution after conversion to HiCN yields an optical density D A D , where D is caused by HiCN and A D by non-HEN. Iron determinations yield a value of c Ac milligram-atom Fe/liter, where c is hemoglobin iron and Ac belongs to non-HiCN iron. From these values the extinction coefficient of HiCN is calculated ( I = 1.000cm) :
+
e =
+
(D
+ A D ) / ( c + Ac)
(5)
If we now introduce e l , the true extinction coefficient of HiCN, keeping in mind that c milligram-atom Fe/liter corresponds to c quarter millimoles Hb/liter, we may change Eq. ( 5 ) to: er =
E
+ (Ac/c)B- ( A D / c )
The true extinction coefficient thus equals the experimentally obtained value plus a correction for non-HiCN iron and minus a correction for nonHiCN extinction. Even if it were possible to evaluate the factor A D / c accurately, it would not be advantageous to apply this correction to the experimentally obtained extinction coefficient. In routine hemoglobin photometry some non-HiCN extinction will always be present. Thus, using an extinction coefficient influenced by a similar factor will at least partially eliminate the non-HiCN extinction error from the determination of hemoglobin concentrations. Theoretically it would be an advantage to apply the correction for nonhemoglobin iron: (Ac/c)c. The determination of the oxygen capacity is not suitable for this purpose, since HbCO and Hi will not be determined. However, these hemoglobin derivatives should certainly be included in a (total) hemoglobin determination. The best method presently available to determine the sum of Hb, HbOz, Hi, and HbCO is the determination of the CO-combining capacity after reduction with NazS2O4.However, as long as the reactions caused by the addition of NazSz04and CO to hemoglobin solutions are not fully understood and no common opinion has been established on the significance of the CO capacity in the presence of NazSzO4, it seems unwise to use this method as a basic procedure for evaluating a%&. Consequently, the factor (Ac/c)e being small but not yet accurately known, no correction should be applied and $&N determined on the basis of total iron should be taken as the true extinction coefficient of HiCN at X = 540 nm. It is therefore obviously justified to adopt iron determination as the basic method for the evaluation of this method being the one with the most solid experimental foundation presently available. The methods
148
E. J. VAN WMPEN AND W. G. ZIJLSTRA
described ih Sections 2.1.2, 2.1.3, and 2.1.4 proved to be suitable for an accurate determination of egkNon the basis of iron. 2.1.2. Spectrophotometric Iron(II) Determination with a,a'-Dipyridyl
Decomposition. A hemoglobin preparation (1.00 ml) is decomposed using 1 ml nitric acid (65%) and 0.5 ml hydrogen peroxide (30%). To prevent foaming three drops of octyl alcohol are added and the solution is carefully heated. A clear, light yellow solution is obtained, which is evaporated almost to dryness. To remove excess acid, 1-2 ml water is added and evaporated in vucuo a t 90-100°C; this is repeated once. The contents of the digestion tube are finally transferred to a volumetric flask and made up to 50 ml with water. Measurement. This solution (2.0 ml) is mixed with 1.0 ml buffered a,a'dipyridyl solution (0.5 g a,d-dipyridyl, 3.4 g sodium acetate trihydrate, and 1.41 ml glacial acetic acid, made up to 50 ml with water) and 1.0 ml sodium sulfite solution (2.52 g Na2S03.7H20 made up to 100 ml with water). After adding 3.5 ml water the mixture is heated for 10 min at 100°C in a water bath. During heating the test tubes are covered with glass marbles and the upper end of the tubes is cooled by an air stream. The optical density of the pink solution is then measured a t room temperature using a Beckman DU or similar spectrophotometer with water as a blank (A = 520 nm; slit width = 0.025 mm; I = 1.000 cm). Appropriate corrections are made for traces of iron in reagents and glassware (blank decomposition) and for the faint yellowish background color of the solution obtained by destruction of the hemoglobin. Calibration. A calibration curve is made using ferric ammonium sulfate, Fez(S04)3-(NH4)2S04.24H20 (analytical grade, Merck), the iron content of which has repeatedly been checked by gravimetric analysis. The salt is stored in an atmosphere corresponding to the water vapor pressure of the crystals (6 mm Hg). A stock solution is prepared containing 1.7272 g ferric ammonium sulfate per liter, which corresponds to a concentration of 3.582 milligram-atom Fe per liter. After 20-fold dilution of the stock solution, 0.5, 1.0, . . . ,and 5.5 ml made up to 7.5 ml with water provide a series of solutions containing 1.194 X 2.388 X . . . , and 13.134 X milligram-atom Fe per liter. The ADKzo between two succeeding steps equals 0.103. As the hemoglobin dilution factor in this procedure is 187.5, the iron content of the hemoglobin sample is obtained in milligram-atom per liter by multiplying the value of DK20by 21.73. 2.1.3. Titrimetric Iron(III) Determination with Titunous Chloride Decomposition. A hemoglobin preparation (1.00 ml) is decomposed using 1 ml concentrated sulfuric acid and 250 mg potassium sulfate. Mercury or
HEMOGLOBIN AND ITS DERIVATIVES
149
selenium cannot be used as a digestion catalyst, because they interfere with the titration. After 30 minutes, heating is interrupted for 5 minutes, after which 1 ml concentrated sulfuric acid and 0.5 ml hydrogen peroxide (30%) are added; the addition of hydrogen peroxide is repeated after 1 and 2 hours. Titration. After cooling, 10 ml water and 1 ml potassium thiocyanate (20%) are added. The titration with titanous chloride is carried out in the digestion tube in a COz atmosphere. The titanous chloride solution (about 0.15%) is frequently standardized against the ferric ammonium sulfate stock solution (mentioned in Section 2.1.2). 2.1.4. Determination of D64O of Hemiglobincyanide
Preparation of HiCN solution. A hemoglobin preparation (1.00 ml) is pipetted into a 250-ml volumetric flask containing about 50 ml diluting solution (200 mg potassium ferricyanide, 50 mg potassium cyanide, and 1.0 g sodium bicarbonate made up to 1liter with water). After homogenization the volume is made up to 250 ml with the same diluting solution. Measurement. The optical density of the HiCN solution is measured after at least 30 minutes using a Beckman DU or similar spectrophotometer with water as a blank (A = 540 nm; slit width = 0.02 mm; 1 = 1.OOO cm). The optical density of the diluting solution a t 540 nm should be zero when measured against water. 2.1.5. Actual Determination of
&& i n 21 Hemoglobin Solutions
To obtain a hemoglobin solution suitable for the determination of &&, about 90 ml fresh human blood is collected in a 200-ml conical flask containing 20 ml 3.2% sodium citrate (Na,CsHb0,.5$H20). The erythrocytes are washed twice with 0.9% NaCl solution. An equal volume of twicedistilled water is added to the erythrocytes, and highly purified toluene is added in an amount of 40% of the original erythrocyte volume. After thorough mixing the fluid is stored at 4°C for 12 hours. After centrifugation the mixture displays three clearly distinct layers. The upper layer consists of toluene, the middle one is a turbid suspension of erythrocyte stromata, and the third is a clear oxyhemoglobin solution. The first two layers are sucked off completely. The HbOz solution is now filtered through ash-free filter paper and stored at 4°C. The egCN was determined for 21 hemoglobin solutions thus prepared using the methods described in Sections 2.1.2,2.1.3, and 2.1.4. All glassware used for the iron determinations was made iron-free as far as possible. Only analytical grade chemicals (Merck) were used. For pipetting hemoglobin solutions, repeatedly calibrated Ostwald pipettes were used. Spectrophotometry was performed with the aid of one Unicam SP-500 and four
E. J. VAN KAMPEN AND W. G . ZIJLSTRA
150
Beckman DU spectrophotometers. The results are presented in Table 1. The values given in the second and third columns were obtained independently in two laboratories; the next column shows the total number of D6'0 determinations in the two laboratories. In most cases the values of D64O determined in the two laboratories are in good agreement. In a few VALUES OF DE&
TABLE 1 BASEDON DIFFERENT DETERMINATIONS OF CHiCN, 21 HEMOGLOBIN PREPARATIONS (Zl)
AND OF ef$&, FOR
a,a'-Dipyridyl
Lab 1
DUQ
Prepsration
TiClr Lab 1
Lab 2
p40
n
p40
n
EMo
n
4
-
-
9 10 7
10.94 11.30 10.93
2 2 5
2 2 1
-
-
-
-
0.3976 0.390 0.331 0.394
8 10 7 9
11.05 11.13 11.10 10.95
8 3 6 9
11.04 11.15 11.08 10.83 10.92 10.94
2 2 2
11.07 11.11 10.95 11.07
3 5 6 3
0.352 0.381 0.398 0.394
0.357 0. 3536 0.399 0.396
8 10 9 9
11.10 10.98 10.89 11.04
6 6 6 4
11.06 11.10 10.92 11.12
2 2 2 2
10.93 11.23 10.97 11.01
6 6 6 6
13 14 15 16
0.350 0.322 0.379 0.376
0.351 0.3226 0.375
11.05 11.04 10.97 10.83
6 6 9 6
11.01 10.81 10.82
-
2 2 2
-
11.19 11.20 10.95 11.08
5 6
-
6 6 7 5
17 18 19 20 21
0,342 0.354 0.321 0.370 0.260
0.3426 0.354 0.320 0.359 0.260
8 9 7 9 6
10.99 10.95 10.98 10.87 11.13
6 9 9 9 6
10.89 10.69 10.80 10.81 11.00
2 2 2 2 2
11.10 11.08 11.01 10.80 11.28
5 9 7 9 6
Lab 1
Lab2
n
1 2 3 4
0.419 0.294 0.419 0.362
0.294 0.420 0.365
5 6 7 8
0.397 0.387 0.330 0.382
9 10 11 12
-
4
9
cases, however, the difference between the values in the second and third columns exceeds the acceptable random error (e.g., preparations 10 and 20). Nevertheless the resulting ~ 2values % ~are in good agreement. The explanation of this fact is to be found in corresponding differences in the results of the iron determinations. In a few preparations a significant difference existed between the concentrations of the solutions as analyzed in the two laboratories. The average values of @kNare presented in Table
151
HEMOGLOBIN AND ITS DERIVATIVES
2, together with the standard deviation and the standard error of the mean. In calculating these quantities, the scatter of the iron determinations as well as that of the spectrophotometric measurements has been taken into account. 2.1.6. Discussion of
as Determined by Various Investigators In the first determinations of &&, made by Drabkin and Austin (D3, D4), hemoglobin concentrations were determined by measuring oxygen capacity using Van Slyke’s manometric procedure; the optical density of the HiCN solutions was measured with a (visual) Konig-Martens spectrophotometer. A value of 11.51 was found as the average of 7 determinations. Spectrophotometry of HiCN solutions was repeated on a larger scale yielding C ~ & N= 11.53 f 0.05 f 0.01 (Dl), but it seems that only the spread of the determinations is of significance in this series, since the average constants had been assumed on the basis of the earlier work. &$N
TABLE 2 AVERAGEAND STANDARD DEVIATIONS~ OF &$,, BASEDON DIFFERENT DETERMINATIONS OF cHiCN Method for determining cHiCN ~~~~
4%N
S
s/n118
1)
~~
Fez+ (a, a’dipyridyl) (lab 1) Fez+ (a, a’-dipyridyl) (lab 2) Fea+(TiCla) (lab 1) s = standard deviation; determinations. 5
10.99 10.94 11.05
s/n1Iz =
0.126 0.153 0.156
0.014 0.026 0.017
123 35 101
standard error of the mean; n = number of
A value of 11.5 for &&N, based on Drabkin’s work, has been widely used as a basis for hemoglobin standardization (Cl, (22). This value has been accepted rather uncritically, in some instances even in the conviction that it was based on hemoglobin iron determinations. This may be partly due to a paper by Lambertsen et aE. (Ll), where the oxygen capacity measurements underlying &‘& = 11.5 are assimilated to hemoglobin iron determinations. In their paper it is stated, “The total pigment was determined independently as cyanmethemoglobin which is equivalent to estimation of hemoglobin iron.” Using a titrimetric iron determination with titanous chloride, Remmer (R3) obtained for &‘& a value of 11.09 as an average of 11 determinations. Minkowslti (M7) found &‘& = 11.15, using a spectrophotometric method involving the use of orthophenanthroline. These values are in fair agreement = with our results (see Section 2.1.5 and Table 2). An indirect proof of E&
152
E. J. VAN KAMPEN AND W.
G.
ZIJLSTRA
11.0 has been provided by Van Oudheusden e2 al. (V4), who compared the HiCN method based on this value for the quarter miUimolar extinction coefficient with a hemoglobin determination involving the evaluation of iron by a modified cu,a’-dipyridyl method. The average difference between the two methods was only 0.024 g/100 ml (n = 17). Very recently Salvati e l a2. (Sl) purified hemoglobin solutions with the aid of a carboxymethyl cellulose (CMC) column, using a phosphate buffer of pH = 6.8 as elution fluid. A modified a,d-dipyridyl method according to Mason and Adarraga-Elisaran (M5) was then used for the determination of iron. ~ 2was % found ~ to be 10.95 (n = 46) with a standard error of 0.03. Thus there now exists overwhelming experimental evidence that the quarter millimolar extinction coefficient of HiCN a t X = 540 nm is very near 11.0. This value was therefore proposed by the standardization committee of the European Society of Hematology a t its 1963 meeting in Lisbon (B2) and was definitely adopted at the next meeting of this committee (Stockholm, 1964, B3). 2.2. THE HEMIGLOBINCYANIDE METHOD 2.2.1. The Diluting Solution To achieve a diluent for hemolysing blood samples as well as for rapid and complete conversion of all hemoglobin derivatives present to HiCN, without the development of turbidity caused by plasma proteins and erythrocyte stromata, the following points should be observed: (1) The absorption spectrum of HiCN, and consequently &%N, is independent of the pH of the solution (V3). (2) The formation of Hi from HbOz, and to a lesser degree of HiCN from Hi, proceeds more rapidly at low pH and low ionic strength (V3). (3) Therefore, and to promote rapid lysis of the red cells, the total amount of solutes in the diluent should be as low as possible. (4) Dilution of blood samples with a solution of near neutral pH often causes a slight turbidity to develop through precipitation of plasma globulins. The consequences of (2) and (4) seem rather incompatible. To acquire a high reaction velocity, a diluent with pH 7 is desirable. However, the use of such a diluent is forbidden became of the inevitable occurrence of slight turbidity. Very different diluting solutions have thus been proposed, depending on the direction of the investigator’s attention. Chilcote and ’ O’Dea (C3), using a pH of 7.4, obtained stable optical density readings within a few minutes, but their results show a signilkant influence of proteins, up to 2.60/,, dependent on the plasma albumin/globulin ratio. Even the conventional diluting solution (mentioned in Section 2.1.4; pH 8.6), may cause globulin precipitation in the case of pathological sera. This
153
HEMOGLOBIN AND ITS DERIVATIVES
persuaded Green and Teal (G3) to add 100 mg KzC03,thus increasing the pH to 8.9. Yet the same result could have been attained more simply by diminishing the amount of NaHC03, the pH of a solution of 200 mg KJ?e(CN)a and 50 mg KCN being 9.6. Using the conventional diluent, stable optical density readings are attained in 15-25 min, which is a distinct disadvantage in clinical practice. Lowering the pH of the diluent to 7.2 yields a stable value for DS4Owithin 5 minutes, irrespective of whether the pH is determined by borate, barbiturate, or phosphate buffers (V3). However, as in the observation of Chilcote and O’Dea, the results are 2-3% too high. Addition of 0.5 ml Sterox SE2per liter diluting solution appeared to suppress the development of turbidity completely, even at pH 7. In this concentration, Sterox SE has no influence on the absorption spectrum of HiCN and increases the over-all reaction velocity by its strong
t
0.400
D5L0
0.390
0.390 -
0.320 0.310min I
I
I
I
I
I
FIG.4. Formation velocity of H E N from HbOz. D 4 0 plotted against time. At zero time 0.5 ml blood was mixed with 99.5 ml of different diluting solutions: Diluting solution containing 200 mg K3Fe(CN)e,50 mg KCN, and 1 g NaHCOa per liter, pH 8.6. Diluting solution containing 200 mg &Fe(CN)@,50 mg KCN, and 140 mg KHiPOd per liter, pH 7.2. - . * - Diluting solution containing 200 mg K S F ~ ( C N )50 ~ , mg KCN, 140 mg KHzP04,and 0.5 ml Sterox SE per liter, pH 7.2.
-
-----
-
2 A polythiol compound manufactured by Hartman-Leddon Company, Philadelphia, Pennsylvania, U.S.
154
E. J. VAN KAMPEN AND W. G. ZIJLSTRA
hemolytic action. Thus a diluting solution of the following composition has been introduced: 200 mg K3Fe(CN)e,50 mg KCN, 140 mg KH2P04,and 0.5 ml Sterox SE made up to 1 liter with water (V3). Figure 4 shows the time course of DK40when using different diluting solutions. The HiCN solution prepared with the conventional diluting solution reaches its final DS4O18 minutes after the moment of mixing blood and diluent. The D640values of the other solutions (pH 7.2) are stable within 3 minutes, the one without Sterox SE giving values slightly too high, caused by turbidity. 2.2.2. Spectrophotometric Procedure
A 2 0 4 homogenized blood sample is added to 5 ml diluting solution. Blood and diluent are mixed well. After 3 minutes the optical density is measured against water (A = 540 nm; slit width = 0.02 mm; 1 = 1.000 cm). The hemoglobin concentration is calculated by means of the equation
c = dfDM0)/tFLNquarter millimoles/liter
(6)
or
c = (fDWM X
lC~-~)/tg&~ g/100ml
(7)
optical density at X = 540 nm, = quarter molecular weight of hemoglobin = 16114. Substituting these values in Eq. (7) gives c = 36.8D"O g/100ml where f
=
dilution factor
=
251, OW
=
&tN= quarter millimolar extinction coefficient = 11.0, and M
2.2.3. Filter Photometric Procedure
In this procedure, a filter with a maximum transmission near 540 nm has t o be used. An interference filter with a half-intensity band width of about 20 nm proved t o be suitable. The hemoglobin concentration is read from a calibration line which is established with the aid of a single HiCN standard solution (Section 2.3). The validity of this procedure is demonstrated in Fig. 5, where it is shown that a calibration line drawn through a point obtained with the standard solution and the origin coincides with a cdibration line established using a series of known HiCN solutions. 2.3. PREPARATION O F A HEMIGLOBINCYANIDE STANDARD SOLUTION A partly purified HbO2 solution is prepared (as described in Section 2.1.5), Ta make up 1 liter of HiCN standard solution, 5 ml HbOz solution is added to the diluting solution (described in Section 2.1.4). The HiCN
155
HEMOGLOBIN AND ITS DERIVATIVES
0.7001 D
c
CHi,,(g/100rnl)
FIG.5. Calibration line of a Beckman C photometer. DS4O of a series of HiCN solutions, measured using an interference filter A(, = 540 nm), has been plotted against the hemoglobin concentration of the undiluted blood; 0 is the result of the measurement of a single HiCN standard solution.
solution thus obtained is sterilized by Seite filtration and transferred, under rigid aseptic precautions, to brown glass ampules of 10 ml. The concentration of the HiCN solution in at least 5 ampules, taken at random, is determined spectrophotometricaly, using ei4$N = 11.0 and a quarter molecular weight of 16114. These determinations are a check on the homogeneity of the lot and yield the value to be printed on the ampules (22). Twice 5 ampules, taken at random, are incubated at 22' and 37°C in four different culture media (broth, glucose broth, ascites broth, and an anaerobic medium) at an interval of some weeks. When no growth occurs the lot is considered to be microbiologically sterile. To check the stability of the preparation, ampules are stored in the dark at 4' and 20°C for at least 1 year. The optical density at X = 540 and 504 nm is measured every month. During this time D ~ $ Nshould not should remain decrease more than 1%, while the quotient Dg&N/D&ycN between 1.62 and 1.58. 3. Absorption Spectra and Extinction Coefficients
When the extinction coefficient of a hemoglobin derivative at a given wavelength is known accurately, any extinction coefficient of any hemoglobin derivative may be determined by using spectrophotometry only. Thus, using E ~ & N= 11.0 (see Section 2.1.5) as a starting point, e/X curves have been determined for Hb, HbOn, HbCO, Hi, and HiCN.
156
E. J. VAN KAMPEN AND W. G . ZIJLSTRA
3.1. SPECTROPHOTOMETRIC TECHNIQUE
A Beckman DU spectrophotometer was used for all measurements. The slit width was chosen so that a half-intensity band width of 1.6 nm was obtained in the region X = 650-590 nm, 0.6 nm in the region X = 590460 nm, and 0.15 nm in the region of the Soret bands. The wavelength of the instrument had been checked with the aid of the mercury emission spectrum. Measurements a t X > 590 nm were made in a layer thickness of 0.0993 cm, a t X < 590 nm in a layer thickness of 0.0128 cm. The latter was obtained by inserting a plane parallel glass plate into the 0.0993-cm cuvette (Fig. 13). The layer thickness of the cuvette had been determined spectrophotometrically using HiCN solutions and a 1.000-cm cuvette as a reference. Assuming Lambert-Beer’s law to be valid for all solutions over the entire spectral region, ex can be calculated from Dx for any hemoglobin derivative, when the total hemoglobin concentration of the sample has been determined; ex = Dx/cl where ex = the quarter millimolar extinction coefficient at wavelength X, Dx = optical density at wavelength X, c = total hemoglobin concentration in quarter millimoles/liter, and 2 = layer thickness in cm. The total hemoglobin concentration was determined using the spectrophotometric HiCN method described (Section 2.2.2) with the exception that a dilution factor of 100 was used, obtained by diluting 1 ml of the hemoglobin derivative solution to 100 ml with the diluting solution. 3.2. PREPARATION OF BLOOD SAMPLES
Fresh blood was obtained from nonsmoking individuals by venipuncture, with heparin added to prevent clotting. The erythrocytes were washed 3 times with isotonic saline solution and the packed cells then brought to the original blood volume with 0.7% Sterox SE solution. After mixing thoroughly, the hemolysate was filtered through a folded paper filter and stored in 20-ml all-glass syringes. 3.3. THEABSORPTION SPECTRUM OF HEMOGLOBIN (HB)
The hemolysate (Section 3.2) was further diluted with 0.7% Sterox SE solution until an optical density of about 0.500, measured in a layer thickness of 0.0128 CM at X = 577 nm, was reached. The total hemoglobin concentration ( C W ~ ~of. )the diluted hemolysate was measured (as described in Section 3.1). To the sample, Na2S204(4 mg/ml) was added for complete conversion to Hb, and the spectrophotometric measurements
157
HEMOGLOBIN AND ITS DERIVATIVES
down to X = 460 nm were made. The diluted hemolysate was further diluted 10-fold with 0.7% Sterox SE, and CHb tot. was determined by adding 90 ml of the diluting solution (described in Section 2.2.1) t o 10 ml of the 150.01 E
FIG.6. ~ / curve k of hemoglobin (Hb). Quarter millimolar extinction coefficients based on ek:EN = 11.0. The soEid part of the graph represents the mean value resulting from 5 samples, the dashed part results from measurements of a single sample. Measurements made with a Beckman DU spectrophotometer, layer thickness 0.0128 cm; 10-fold dilution of samples for measuring below A = 460 nm.
sample and measuring at X = 540 nm. NazSzOr (0.4 mg/ml) was then added to the sample, and the measurements down t o X = 400 nm were carried out. The resulting €/A curve is shown in Fig. 6. TABLE 3
c
VALUES OF LIGHTABSORPTIONMAXIMAAND MINIMAOF HB
Wavelength (nm)
(n = 5)
Mean E
556-554 47s 431
13.04 3.31 140.10
Highest value 13.26 3.45 141.70
A%
+1.7 +4.2 +1.1
Lowest value 12.86 3.17 138.44
A% -1.4 -4.2 -1.2
Table 3 presents the extremes of the c values found for the light absorption maxima and minima; in Table 4 some of the isobestic wavelengths and corresponding e values are given. It is interesting to note the agreement of our value for 8;and that found by Benesch et al. (Bl), who added NADHz (formerly called DPNH), methylene blue, and methemoglobin reductase 2 = 140.0 based on ek&~= to their hemoglobin solution and arrived at ; 11.5.
158
E. J. VAN KAMPEN AND W. G . ZIJLSTRA
TABLE 4 ISOBESTIC POINTS(IBP) OF HB WITH &ot, B C O , AND HI; E VALUEB OF ISOBESTIC WAVELENQTHS Hemoglobin IBP derivative (nm) HbO, HbCO Hi
E
586 7.23 579 8.86 600 3.20
IBP (nm)
E
569 11.27 561.5 12.54
-
-
IBP (nm)
E
548.5 12.46 547.5 12.37
-
-
IBP (nm)
E
6.42 522 528.5 7.71
AT THE
IBP (nm)
e
507 4.81
- - -
3.4. THEABSORPTION SPECTRUM OF OXYHEMOGLOBIN (HBO~)
The hemolysate (Section 3.2) was further diluted with 0.7% Sterox SE solution until an optical density of about 0.400, measured at h = 577 nm in a layer thickness of 0.0128 cm, was reached. Of this solution 10 ml was brought into a revolving glass tonometer which was 30-cm long. The tonometer was continuously flushed with pure oxygen (1 liter/min) for
FIG.7. c/X curve of oxyhemoglobin (HbOZ). Quarter millimolar extinction coefficients based on = 11.0. The solid paTt of the graph represents the mean value resulting from 5 samples, the dashed part results from measurements of a single sample. Measurements made with a Beckman DU spectrophotometer, layer thickness 0.0128 cm; 10-fold dilution of samples for measuring below X = 460 nm.
EZ&
10 minutes, C Htot. ~ was determined (as described in Section 3.1), and the spectrophotometric measurements down t o h = 460 nm were made. The HbOz solution was next diluted 10-fold with Sterox SE 0.7%, the C Htot. ~ was redetermined (as described in Section 3.4), and the measurements
159
HEMOGLOBIN AND ITS DERIVATIVES
TABLE 5 VALUESOF LIGHTABSORPTIONMAXIMA
c
_ _ _ _ ~
~
~
AND
MINIMAOF H B O ~
__
~
Wavelength (nm)
Mean c (n = 5)
Highest value
A%
Lowest value
A%
577 560 543-541 510 415
15.37 8.47 14.37 4.76 130.98
15.46 8.51 14.44 4.80 133.89
$0.6 +0.5 +0.5 $0.8 $2.2
15.31 8.40 14.27 4.68 127.30
-0.4 -0.8 -0.7 -1.7 -2.4
TABLE 6 HBO~ WITH HB, HBCO, AND HI; THE ISOBESTIC WAVELENGTHS
ISOBESTIC POINTS(IBP) Hemoglobin IBP derivative (nm)
OF
IBP (nm)
c
IBP (nm)
-
IBP (nm)
e
-L
\;o
,.40.0
II
5.00
660
e
II
/
3.00 1.00 -
VALUESOF H B O ~AT
IBP (nm)
200.0
111.00 13.00 7.00 9.00 5 ' 0 -0 i t
e
;
LO.0
I
: 610
590
'\
/
-
20.0-
I'
,'
----f' ,
,/\\' I
I!
630
-
,\
I
I
I
A (nm)
, 570
5O ;
5iO
510
490
L;O ' -So L;O
4;O
&
FIG.8. c/X curve of carboxyhemoglobin (HbCO). Quarter miilimolar extinction coefficients based on ck%N = 11.0. The solid part of the graph represents the mean value resulting from 5 samples, the dashed part results from measurements of a single sample. Measurements made with a Beckman DU spectrophotometer,layer thickness 0.0128 cm; 10-fold dilution of samples for measuring below A = 460 nm.
160
E. J. VAN KAMPEN AND W. G. ZIJLSTRA
down to X = 400 nm were carried out. The resulting e/X curve is depicted in Fig. 7. Table 5 presents the extremes of the e values found for the light absorption maxima and minima; in Table 6 some of the isobestic wavelengths and = 132, corresponding e values are given. Benesch et al. (Bl) found based on eggN= 11.5. 3.5. THEABSORPTION SPECTRUM OF CARBOXYHEMOGLOBIN (HBCO) The HbOz hemolysate described in Section 3.4 was used. Tonometry of a 10-ml sample was carried out for 15 minutes while continuously flushing with 99% pure CO, and the CHbtot. was determined (Section 3.1). I n this case, measurement of DEgN was not made for at least 2 hours, as the conversion of pure HbCO t o HiCN proved to take some 90 minutes. The spectrophotometric measurements were then made down to X = 460 nm. For the measurements down to h = 400 nm, a further 10-fold dilution of the sample and a redetermination of CHI,tot. (Section 3.3) were carried out. The resulting e/h curve is given in Fig. 8. 6
TABLE 7 VALUESOF LIQHTABSORPTION MAXIMA AND MINIMA OF HBCO
(4
(n = 5)
Highest value
A%
Lowest value
A%
569-568 555 539 496 420
14.36 11.33 14.33 5.16 191.91
14.43 11.45 14.37 5.27 198.52
+0.5 +1.1 +0.3 +2.1 $3.4
14.32 11.25 14.30 5.07 184.32
-0.3 -0.6 -0.2 -1.7 -4.0
Wavelength
Mean e
The extremes of the e values found for the light absorption maxima and minima are given in Table 7; some of the isobestic wavelengths with their corresponding t values are given in Table 8. ISOSESTIC POINTS(IBP) Hemoglobin IBP derivative (nm) Hb HbOa Hi
OF
t
579 8.86 572.5 13.50 585.5 3.79
TABLE 8 HBCO WITH HB, HBO~,AND HI; c VALUESOF HBCO AT THE ISOBESTIC WAVELENQTHS IBP (nm)
e
561.5 12.54 549.3 12.06
-
-
IBP (nm)
6
547.5 12.37 540 14.27
-
-
IBP (nm)
e
-
-
-
519.7 8.02
IBP (nm)
e
-
-
497 5.16
-
-
161
HEMOGLOBIN AND ITS DERIVATIVES
I
90
Y A (nm)
FIG.9. Light absorption spectra of hemiglobin (Hi) a t varying pH. Preparation of Hi from HbOz by meana of excess &Fe(CN)a. Recorded with a Beckman DB spectrophotometer, layer thickness 0.007 cm. pH 9.12 by means of 0.2 N NaOH; ---- pH 8.06 by means of 0.2 N NaOH; - ..- pH 7.02 by means of 0.5 N lactic acid; -pH 5.94 by means of 0.5 N lactic acid. - a
0 .
160.0
120.0 1LO.O
p E
100.0
7.00 9.00
-c
5.00
-
3.00 #--,
-/'
,'
*/
I0
8a0c-*!
66.0
LO.0
we*--*
I
1.00
II
-
I
I
I
/I
//
20.0 0
--
!,
II
0
A (nrnl
FIG. 10. €/A curve of hemiglobin (Hi). Quarter millmolar extinction coefficients based on = 11.0. The solid part of the graph represents the mean value resulting from 5 samples, the dashed part results from measurements of a single sample. Measurements made with a Beckman DU spectrophotometer, layer thickness 0.0128 cm; 10-fold dilution of samples for measuring below A = 460 nm.
eFcCN
162
E. J. VAN KAMPEN AND W. G . ZIJLSTRA
3.6. THEABSORPTION SPECTRUM OF HEMIGLOBIN (HI)
Although, as can clearly be seen from Fig. 9, the light absorption spectrum of Hi is greatly dependent upon the pH, no specific measures were taken to ensure all samples having the same pH. The hemolysate (Section 3.2) was further diluted with an equal volume of 0.7% Sterox SE solution, and 100 mg finely powdered KaFe(CN), was added to 10 ml of the mmple. After some 15 minutes a further 10 mg K3Fe(CN),was added to give an end molar ratio of hemoglobin/&Fe(CN) 6 of approximately 1:8. The pH was then measured using a radiometer PHM-4C with a capillary glass electrode. Solutions in the pH range 7.0-7.4 were judged suitable and measured spectrophotometrically down to X = 460 nm after determination of CAI, (Section 3.1). A further 10-fold dilution of the sample was made, Cmt,,t. was redetermined (Section 3.3), and the measurements down to X = 400 nm were carried out. The resulting e/X curve is shown in Fig. 10. In Table 9 extremes of the e values for the light absorption maxima and minima are presented. In this small series no correlation between the t
TABLE 9 VALUESOF LIGHTABSORPTION MAXIMAAND MINIMAOF HI
Wavelength
(4 630 610 502-500 476 406
(n = 5)
Highest value
A%
Lowest value
A%
3.69 3.10 8.93 7.97 161.22
3.85 3.22 9.07 8.28 167.48
$4.3 $3.9 +1.6 +3.9 +3.9
3.44 2.95 8.62 7.78 153.21
-6.8 -4.8 -3.5 -2.4 -5.0
Mean t
pH in this range and the resulting t values could be found. Table 10 gives some of the isobestic wavelengths together with the e values. TABLE 10 ISOBESTIC POINT8 (IBP) OF HI WITH HB, HBO~,AND HBCO; t VALUES ISOBESTIC WAVELENGTHS Hemoglobin derivative
IBP (nm)
c
Hb HbOz HbCO
600 -
3.20 -
-
-
OF
HI
AT THE
IBP (nm)
c
IBP (nm)
c
590.7 585.5
3.62 3.79
528.5 525.3 519.7
7.71 7.72 8.02
The c values for the Soret band may have been influenced by the presence of porphyrins (see also Section 3.8). Light absorption by &Fe(CN), in
163
HEMOGLOBIN AND ITS DERIVATIVES
this region is negligible because of the relatively small amount and the extremely thin layer used.
3.7. THEABSORPTION SPECTRUM OF HEMIGLOBINCYANIDE (HICN) Different HiCN standard solutions, prepared by the Dutch Institute of Public Health (Utrecht, The Netherlands) for the calibration of photometric hemoglobin determinations, were measured in a layer thickness of 1.000 cm. The hemoglobin concentration was calculated from the Dfi40 reading. For the measurements in the spectral region 460 > X > 400 nm, a layer thickness of 0.0993 cm was used. AS a check, a HiCN sample was prepared by diluting the hemolysate of Section 3.2 with 0.7y0Sterox SE until an optical density of 0.500 at X = 577 nm (layer thickness 0.0128 cm) was reached, and then adding 200 mg K3Fe(CN), and 50 mg KCN to 10 ml of this solution. The CWtot. was determined (as described in Section 3.1) as if the solution contained some unknown hemoglobin sample. The spectrophotometric measurements were then performed in layer thicknesses of 0.0993 and 0.0128 cm down to X = 600 and 460 nm, respectively. For the measurements down to X = 400 nm the solution was further diluted 10-fold with 0.7% Sterox SE, and the C H b tot. was redetermined (Section 3.3). The resulting e values did not differ by more than 0.09 when compared with the mean values of the first series of determinations for the spectral region down to X = 460 nm. The differences a t X = 540 and 504 nm were -0.03 and 0.00, respectively. The differences in the region of the Soret band were less then 4% of the values of the first series, all values of the last sample being lower. This difference may well be due in part t o light absorption by K3Fe(CN)e. The optical density of all samples in the near infrared was also measured. Whereas the first series shows no absorption in this region, the freshly prepared sample does to a slight degree, undoubtedly caused by incomplete elimination of erythrocyte stromata from the sample. The results are given in Table 11. TABLE 11 LIGHTABSORPTION OF HICN SOLUTIONS IN ~~
Preparation R.I.V.a (n = 5) Concentrated sample
tlOOO
0.02 0.10
THE
NEARINFRARED
~~
e900
,800
0.00 0.05
0.00 0.03
,700
0.03 0.03
p
6
0.10 0.12
Rijks Instituut voor de Volksgezondheid (Dutch Institute of Public Health).
The mean e/X curve from the first series is presented in Fig. 11. The extremes of the t values for the light absorption maxima and minima,
164
E. J. VAN KAMPEN AND W. G. ZIJLSTRA
14.00 16.00
12.00
-
10.00
-
8.00
-
t
6.00 -
650
630
610
590
570
550
530
510
190
L70
460
LLO
120
LOO
PIG. 11. e/X curve of hemiglobincyanide (HiCN). Quarter millimolar extinction coefficients based on ei:EN = 11.0. Results shown are the mean of 5 samples. Measurements made with a Beckman DU spectrophotometer, layer thickness 1.000 cm down to X = 460 nm, 0.0993 cm from = 460 nm down to X = 400 nm.
together with the value of the quotient sented in Table 12. e VALUES OF
e 5 4 0 / ~ 5 0 4 (see
Section 2.3), are pre-
TABLE 12 LIGHTABSORPTIONMAXIMA A N D MINIMAO F HICN @40
/e
AND THE VALUE OF
604
/e
Wavelength Mean t Highest (n = 5) value (nm) 542-540 504 421
11.0 6.83 6.87 122.50 125.64
A%
Lowest value
-
-
+0.6 +2.6
6.79 120.19
604
A%
High
Low
-0.6 -1.9
1.614 -
1.606
-
3.8. LIGHTABSORPTION BY NON-HEMOGLOBIN
In the determination of the quarter millimolar extinction coefficients of hemoglobin derivatives, the following sources of error are present : (a) Light absorption by the chemicals used. The absorption by NazSz04, KCN, and K3Fe(CN)6 in the spectral regions studied has been checked. Only K3Fe(CN)6 proved to have a possibly disturbing effect as it shows an absorption peak at X = 420 nm, the millimolar extinction coefficient being 1.01. Because of the quantities used and the layer thickness in which measurements were performed (0.0128 em), the effect upon the calculated
165
HEMOGLOBIN AND ITS DERIVATIVES t
values for hemoglobin derivatives is negligible, except in the case of the
HiCN standard solutions measured. (b) Light absorption by porphyrins. I n the context of the measurements made here the possibility of a disturbing effect through the presence of coproporphyrin 111 was considered. In Fig. 12 the t/X curves of the dif-
l6.m t 14.00 12.00 10.00 8.00 6.00
400 2.00
650
sjg
By)
590
570
550
530
510
LW
470
460
440
420 Loo A h )
FIG.12. t/X curves of hemoglobin derivatives, based on a quarter millimolar extinction coefficient of 11.0 for HiCN a t x = 540 nm. Measurements made with a Beckman D U spectrophotometer, layer thickness 0.0128 cm. For the measurements below X = 460 nm, 10-fold dilution of the samples. Half-intensity band width 1.6 nm for 650 nm < X < 590 nm, 0.6 nm for 590 nm < < 460 nm, and 0.15 nm for 460 nm < X < 400 nm. - .. - .. = Hb. ~- = HbO..- 1 - - - - - = HbCO; -- - - = Hi (pH 7.0-7.4); -. -. - = HiCN; - ... - ... = coproporphyrin 111, pH 7.0; -- .. - - = coproporphyrin 111, pH 7.7.
.
ferent hemoglobin derivatives studied are shown together. In addition the millimolar t values found by Zondag and van Kampen (24) for coproporphyrin I11 a t pH 7.0 and 7.7 have been drawn in for the spectral region X = 430400 nm. It is clear that, if present, a disturbing influence can be expected only for the E values of the Soret band of Hi. As hemolyzed thricewashed erythrocytes were used in this study, however, it is improbable that any quantity of coproporphyrin was present in the solutions. 4. General Aspects of Two-Wavelength Methods
The determination of the optical density of a two-component system a t wavelengths XI and Xz provides two equations:
166
E. J. VAN KAMPEN AND W. G. ZIJLSTRA
Dividing DX1by DX2yields
The total concentration of light-absorbing substances c is introduced, and c - c2 is substituted for cl. If, furthermore, Xz is an isobestic wavelength, then 42 = &'. These conditions being fulfilled, Eq. (8) may be transposed to DX' +'c ($' - c:')cZ _ DX2 eyC
+
The concentration of component 2 is now expressed as a fraction of the total pigment concentration, and the equation is solved for c2/c: ci2 DX1 _ -- cX1 ~ - - -- X 1 - eX1 DX2 2 1 2 5
c2
or C ~ / C=
a(DX'/DX2) -b
(9)
The constants a and b in Eq. (9) can be calculated from a series of measureof solutions containing exclusively component 1 or comments of DX1/DA2 ponent 2. This gives the equations 0
=
a[DX'/DX2]1 -b
1
=
a[DX1/DX2]2 -b
and
[DX1/DX22]1 and [DX1/DX2]2 being determined, a and b follow from these equations. B, the relationship between cz/c Equation (9) is of the form y = Ax and DX1/DX2 thus being linear. Knowing two points only, it is therefore possible to construct a calibration line. If, instead of an isobestic wavelength, any other nonisobestic wavelength is chosen, the relationship is nonlinear: between c2/c and DX1/DX2
+
c2
HEMOGLOBIN AND ITS DERIVATIVES
167
which is of the form
’
=
A - x A - B - (1 - C)z
It is clear that in this case three constants must be determined. If the components 1 and 2 are used only for calibration, extreme care must be taken, when converting 1 to 2, that the total concentration is not altered in any way. If this is not the case, dangerous errors will be made in the determination of the constants B and C , for replacement of E and D in Eq. (10) is allowed only when this requirement has been fulfilled. When use is made of the E values for hemoglobin derivatives (given in Section 3) to overcome this difficulty, one must be certain that measurement of the D values of the sample is performed under exactly the same conditions a s those under which the E values were determined. Any calculation involving Eqs. (9) and (10) presupposes a two-component system. It must be emphasized that this condition is usually fulfilled only as a good approximation. This obviously limits the accuracy of every two-wavelength method. The necessity of working in a two-component system often complicates the technique insofar as certain measures must be taken to arrive at the desired two components. In the determination of SCO,for example, it is necessary either to saturate the sample with oxygen or t o reduce the nonHbCO t o H b by means of Na2S204.In both cases errors may be introduced (Section 6.2.2). 5. Spectrophotometric Determination of the Oxygen Saturation
Many two-wavelength techniques for the determination of the oxygen saturation (Sa) have been evolved. As little light is transmitted by H b and HbOz solutions below X = 600 nm, methods employing red and nearinfrared light are in use (Fl, G1, G2, Jl). Although excellent results seem to have been procured with these methods, two fundamental objections can be raised. The isobestic wavelength around X = 800 nm is quite elusive, not being always found. Also the light absorption of non-hemoglobin in the red and near infrared is relatively large, introducing a n important source of error. For this reason many workers have sought ways to overcome the difficulties encountered when measuring in the spectral range X = 600500 nm. As the obvious answer to the problem, diluting the samples, is not permissible in the case of the So, determination, special cuvettes with very thin layers have been constructed (D5, H2, K1, N l ) . Two of these cuvettes have been extensively tested by Refsum and Sveinsson (R2). They compared the results of oxygen saturation determinations using these cuvettes
168
E. J. VAN KAMPEN AND W. G . ZIJLSTRA
with those obtained by Van Slyke analysis and found good agreement. In our experience the simple solution of using a 0.1-cm cuvette, into which a plane parallel glass plate of the desired thickness is inserted after filling the cuvette with the sample (Fig. 13), gives excellent results. In this way layers down t o 0.005 cm can be procured without difficulty.
FIG. 13. Filling a 0.10-cm cuvette and inserting a 0.09-cm glass plate. A: Cuvette holder, 0.10-cm cuvette, 0.09-cm glass plate; the syringe contains the blood sample. 0: Blood is syringed into the cuvette. C: The glass plate is inserted, leaving a blood layer of 0.01 cm. D: The cuvette is placed in the cuvette holder, which already contains a similar cuvette with distilled water.
Although Refsum (Rl) has demonstrated that very good results can be achieved using X = 576 and 560 nm (nonisobestic), the methods described (Section 5.1.1) have been chosen because the use of an isobestic wavelength offers distinct advantages (Section 4). 5.1. DETERMINATION OF SO,MEASURED AT X = 560 AND 522 h = 560 AND 506 NM, RESPECTIVELY; I = 0.01 CM
NM, AND
Blood (0.5 ml) is transferred anaerobically from the glass syringe containing the sample to a 2-ml syringe. The syringe has previously been filled with 0.5 ml of a n oxygen-free 2% Sterox SE solution and a metal mixing
169
HEMOGLOBIN AND ITS DERIVATIVES
ring. After thorough mixing and discarding the first three drops, the 0.10-cm cuvette is filled. A 0.09-cm plane parallel glass plate is inserted into the cuvette, and the optical density at 560, 522, and 506 nm is measured. The oxygen saturation is calculated from either
So, (%) = [3.08 - l.62(D6so/D622)]100
(12)
[3.13 - 1.22(D660/D606)]100
(13)
or
SO,(%)
=
The constants 3.08 and 1.62 have been calculated from a series of 25 measurements of D660/D622 for 0% HbOz and 100% HbOz; the constants 3.13 and 1.22 from a series of 17 measurements of D660/D606 for 100% HbOz and 11 measurements for 0% Hb02. The series of 17 measurements (100% Hb02) had a mean value of D660/D606 = 1.753 with a standard deviation of 0.008; the series of l l measurements (OyoHbOz) had a mean of D660/D606 = 2.575 with a standard deviation of 0.012. To get a n impression of the accuracy which may be expected when using these constants, 1.753 f 0.008 and 2.575 f 0.012 have been substituted in Eq. (13); 1.753 f 0.008 gives XO., = 100 f 1.0%, and 2.575 f 0.012 gives SO,= 0 f 1.5%. OF THE Two METHODS 5.2. COMPARISON
In a series of 30 blood samples sent to the clinical laboratory for routine arterial Sol determination, the saturation was determined using both the TABLE 13 COMPARISON OF THE Two METHODS DESCRIBED IN SECTION 5.1 so,
88 93 85 94 98 95 100 96 98 99 95 95 94 100 96
I
(%)
87 93 83 96 99 95 97 95 98 100 96 97 91 100 97
so, (%I 69 87 94 95 94 85 94 88 96 95 95 88 96 97 95
59 87 95 90 92 83 95 89 96 95 96 92 100 98 95
170
E. J. VAN KAMPEN AND W. G . ZIJLSTRA
X = 560 and 522 nm and the X = 560 and 506 nm method. The results are given in Table 13. I n this table each value of So, is the mean of a duplicate determination. The standard deviations of the duplicate determinations have been calculated for both methods. Using X = 560 and 522 nm a standard deviation of 1.5% saturation was found; in the case of X = 560 and 506 nm the standard deviation was 1.4% saturation. The isobestic points were established at X = 522.2 f 0.2 nm (n = 100) and at X = 506.0 f 0.4 nm (n = 15), respectively. Finally, when comparing the two methods, the standard deviation of the difference between the corresponding saturation values (Table 13) was calculated at 1.4% saturation. I t should be borne in mind that establishment of the isobestic point at X = 522 nm should be performed extremely carefully, as both the e/X curves of H b and of HbOz are quite steep in this region (Fig. 12). Accurate wavelength calibration of the spectrophotometer used is thus a n absolute requirement,
5.3. COMPARISON OF Son MEASURED SPECTROPHOTOMETRICALLY AT X = 560 AND 522 NM AND DETERMINED BY VAN SLYKE’S MANOMETRIC METHOD In a series of 30 blood samples taken during heart catheterization the SO, was determined using Van Slyke’s manometric technique and the two-wavelength method using X = 560 and 522 nm. The results are given in Table 14. The mean difference between the spectrophotometric and manometric methods was +0.57Oj, saturation with a standard deviation of 2.8% saturation. 6. Spectrophotometric Determination of Carboxyhemoglobin
As long as carbon compounds are used for the production of energy and in heating, carbon monoxide intoxication will remain a danger to man. Besides the fact that hemoglobin has a strong affinity for CO (about 200 times the affinity for O,), the presence of HbCO in blood also impairs oxygen transport by causing a significant shift to the left of the oxygen dissociation curve. This phenomenon has been described as partial molecular poisoning (D2), since the underlying cause is an increase in the oxygen affinity of the remaining iron atoms when a single Fe-CO bond is present in the Hb, molecule. This fully explains why a relatively low HbCO concentration is accompanied by severe symptoms, 25-30% sufficing t o cause unconsciousness. Through the resulting anoxia, HbCO intoxication can give widespread disorders of the nervous system (polyneuritis), even demyelinization within the central nervous system having been found.
171
HEMOGLOBIN AND ITS DERIVATIVES
TABLE 14 COMPARISON OF SO, AS MEASURED BY VAN SLYKE ANALYSISAND SPECTROPHOTOMETRIC MEASUREMENT AT X = 560 AND 522 NM
BY
s o * (%)
61.6 83.5 73.3 92.3 72.5 72.8 88.8 88.0 99.1 75.9 91.4 83.2 94.8 53.0 73.5 42.3 59.6 67.8 33.9 68.8 89.5 79.6 93.7 57.2 73.2 84.0 54.8 71.4 67.8 89.4
59.7 81.1 72.4 88.7 74.4 76.1 93.3 87.3 98.8 78.3 95.6 83.0 94.9 53.1 78.6 41.3 58.4 73.1 34.9 71.2 91.5 82.9 96.0 55.2 73.0 89.0 49.1 70.5 64.6 88.1
-1.9 -2.4 -0.9 -3.6 +1.9 +3.3 +4.5 -0.7 -0.3 $2.4 +4.2 -0.2 $0.1 +0.1 +5.1 -1.0 -1.2 +5.3 $1.0 $2.4 +2.0 +3.0 +2.3 -2.0 -0.2 +5.0 -5.7 -0.9 -3.2 -1.3
In normal blood 0.24.5% HbCO is present, due to CO formation from the a-methene groups of the hemoglobin molecule during physiological decomposition (M2, 52). In smokers up to 3% HbCO has been found. The frequent occurrence of carbon monoxide poisoning necessitates the availability of rapid and accurate methods for the determination of HbCO in human blood samples. These methods may also be of value in evaluating the results of the usual oxygen therapy in severely poisoned patients, and in the detection of cases of chronic poisoning of moderate degree. A review
172
E. J. VAN KAMPEN AND W. G. ZIJLSTRA
of various, essentially spectrophotometric methods is given in references (Vl) and (V2). In this section, 2 two-wavelength methods making use of an isobestic point, which have proven to be quite suitable in our laboratories for many years, are described, The general requirements dealt with in Section 4 are also valid for these techniques. 6.1. DETERMINATION OF Sco IN A MIXTURECONTAINING HB AND HBCO, AT X = 538 AND 578 NM; 2 = 1.000 CM MEASURED 6.1.1. Procedure
Blood (0.1 ml) is mixed with 20 ml 0.1% ammonia solution, and 20 mg sodium dithionite (NarS204)is added to convert HbO2 to Hb. The optical density is measured a t X = 538 and 578 nm (isobestic wavelength); 1 = 1.000 cm; slit width = 0.02 mm. The measurements should be carried out within 10 minutes of the addition of Ka2S204.The CO saturation is calculated from the equation Sco
(70)= [(2.440538/0578) - 2.68]100
(14)
The constants 2.44 and 2.68 have been calculated from a series of 30 measfor 0% HbCO (100% Hb) and 100% HbCO. The urements of D538/D57* isobestic point was established at X = 578 f 0.5 nm ( n = 100). 6.1.2. Calibration Normal blood (0.1 ml) is mixed with 20 ml 0.1% ammonia solution and divided into two equal lots. To each lot 20 mg Na2S204is added to arrive at 100% Hb. Through one of the lots pure CO, formed by heating HCOOH and H2SO4,is allowed to bubble slowly for 2 minutes to give a 100% HbCO
oL
;o
o;
o;
a,
o;
$0
io
00
9’0
So(%)
FIG.14. Calibration line for the determination of SCO,based on Eq. (14).
173
HEMOGLOBIN AND ITS DERIVATIVES
solution. Both samples are then measured at X = 538 and 578 nm; I = 1.OOO cm; slit width = 0.02 mm. The measurements should be carried out within 10 minutes of the addition of Na2S2O4. In a series of 30 determinations, [ D 6 3 8 / 0 6 7 8 ] H b equaled 1.10 f 0.02 and [ ~ 6 3 8 / ~ 6 7 s ] H t , C 01.50 f 0.03. I n practice it proved advantageous to make has been plotted against SCO. use of a calibration line. In Fig. 14, D638/D678 The HbCO saturation of any blood sample can be easily read from this graph. 6.1.3. Discussion
It has been stated ( P l ) that a certain amount of CO dissociates from HbCO when the sample is highly diluted. This view is based on the finding that a change in photometric reading is observed when CO is allowed t o run through a diluted HbCO solution containing Na2SzOs. As the method described (Section 6.1.1) involves a dilution factor of 200, the values of SCO should therefore be too low. It has, however, been shown (Vl) that a change in photometric reading also occurs when using hydrogen instead of CO, or when a dilute HbCO solution is simply allowed to stand for 30-50 minutes in the presence of Na2S204.This has led to the supposition that compounds 0.600
0.500
t
D
-
0
A (nm) 1
660
1
640
1
1
620
1
1
600
1
1
580
1
1
560
1
1
540
FIQ.15. Effect of running CO for 2, 6, and 10 min through a dilute HbCO solution containing Na2P201.__- 2 min CO; - - - - 6 min CO; - . -. 10 min CO.
174
E. J. VAN KAMPEN AR’D W. G . ZIJLSTRA
other than HbCO are formed under these conditions. An extensive spectrophotometric investigation (V2) revealed that, after dispersing CO in a Hb solution containing NazSz04, an increasing amount of SHbCO, sulfcarboxyhemoglobin, may be formed. In Fig. 15 are shown the absorption spectra of a HbCO solution after 2, 6, and 10 minutes of bubbling CO through the solution. In the absorption spectrum the appearance of a peak with a maximum a t X = 614 nm, characteristic for SHbCO (N2), can be observed. The formation of SHbCO is a distinct disadvantage of the procedure described in Section 6.1.1. When a HbCO sample is prepared in the calibration procedure by dispersing CO in a dilute Hb solution containing NazSz04, formation of small amounts of SHbCO is almost inevitable. From the results presented (Section 6.3) it may, however, be concluded that only slight errors are thus introduced. 6.2. DETERMINATION OF SCOIN A MIXTURECONTAINING HBO~ AND HBCO, MEASURED AT X = 562 AND 540 NM; I = 0.01 CM 6.2.1. Procedure
To convert Hb to HbOz, a heparinieed blood sample is rotated for 5 minutes in a small tonameter filled with room air. The CO saturation of the sample is not influenced measurably by this procedure. The sample is then diluted with an equal amount of 2% Sterox SE solution in a syringe containing a metal mixing ring. After thorough mixing and discarding the first three drops, the 0.10-cm cuvette is filled. A 0.09-cm plane parallel glass plate is inserted into the cuvette, leaving a blood layer of 0.01 cm (Fig. 13). The optical density is measured a t X = 562 and 540 nm (isobestic wavelength). The CO saturation is calculated from the equation Sco
(%) = [(3.260662/D640) - 1.98]100
(15)
The constants 3.26 and 1.98 have been calculated from a series of 10 measfor 0% HbCO (100% HbOz) and 100% HbCO. urements of Db62/Db40 6.2.2. Discussion
When measuring in the system HbOz/HbCO, the use of a layer thickness of 1.000 cm is impossible without diluting the blood sample some 100 times. In such circumstances, however, a conversion of HbCO to HbOz would take place because of the oxygen-rich environment. The result of dilution of a sample containing HbCO and HbOz is clearly demonstrated in Table 15.
175
HEMOGLOBIN AND ITS DERIVATIVES
EFFECTON Sco
WHEN
TABLE 15 DILUTINQ A BLOOD SAMPLE CONTAINING HBCO A N D HBO~ ~~~
Dilution factor
I ayer thickness (cm)
s c o (%)
Undiluted 2.5 10 100 200
0.013 0.013 0.100 I ,000 1.000
34 33 32 30.5 29.5
It is therefore necessary to use cuvettes with very thin layers; although this may seem complicated, these cuvettes (23) are in fact very easy to handle. When measuring in thin layers with a dilution factor 1, little or no conversion of HbCO to HbOz is encountered. To demonstrate this, three samples containing HbCO were measured as follows: (a) system HbCO/Hb, addition of R’azS204(method 6.1.1) (b) system HbCO/Hb, eaturation with 0 2 then addition of NanS20n(method 6.1.1) (c) system HbCO/HbOz, saturation with On (method 6.2.1)
The results are present,ed in Table 16. TABLE 16 EF~ECT OF SATUR.4TION WITH OXYGEN ON sco AS DETERMINED WITH IN SECTIONS 6.1.1 A N D 6.2.1 DESCRIBED System
Method
HbCO/Hb HbCO/Hb HbCO/HbOz
6.1.1 6.1.1 6.2.1
HbCO/Hb HbCO/Hb HbCO/HbOp
6.1.1 6.1.1 6.2.1
HbCO/Hb HbCO/Hb HbCO/HbOz
6.1.1 6.1.1 6.2.1
Saturation with 0 2
-
+ + + + + +
THE
METHODS
SCO
(%)
32 32 32 19 21 20 18 20 1s
The method given in Section 6.2.1 may be regarded as the most suitable spectrophotometric method for the determination of SCO.If, however, proper measures are taken against errors caused by the formation of SHbCO, the procedures 6.1.1 and 6.2.1 are of the same order of accuracy. In the calibration experiments of procedure 6.2.1, the average value of [D662/D640]HbOr was 0.609 with a standard deviation of 0 . 0 2 . For a group
176
E. J. VAN KAMPEN AND W. G . ZIJLSTRA
of 16 nonsmokers, however, [0662/0640]HbOa was found t o be 0.603 with a standard deviation of 0.004.Use of Eq. (15) may therefore well give slightly negative results in a few cases. 6.3. COMPARISON OF
THE
Two METHODS
Fresh normal heparinized blood (25 ml) was hemolyzed with 3 ml 7% Sterox SE solution; 10 ml of the hemolysate was rotated for 3 minutes in a small tonometer filled with oxygen to procure a solution containing 100% HbO, (0662/0640 = 0.609). The H i content of this solution was estimated to be near zero with the method described in Section 7.3. This solution was designated A. Another 10 ml of the hemolysate was rotated for 5 minutes in a small tonometer with pure CO to procure a solution containing 100% HbCO (D662/0640 = 0.915). This solution was designated B. From A and B, mixtures were made containing approximately 40,35, 30, 25, 20, 15, 10, 5, and 2% HbCO, respectively. The HbCO concentrations were determined according to the two methods described. The results, given in Table 17, show very good agreement. TABLE 17 SCOMEASURED BY METHODS 6.1.1 AND 6.2.1 IN MIXTURES OF SOLUTIONS CONTAININQ 100% H B O ~AND 100% HBCO, RESPECTIVELY Sco
(%I
Calculated
Method 6.2.1
Method 6.1.1
100 40 35 30
ion
40
25
20 15 10 5
2 0
40 35 29 22 18 13 10 4 1 1
3i 32
25 17 13 8 5
2 0.5
The reliability of the two methods was further checked by determining the SCO of a series of blood samples brought into contact with varying amounts of CO (Table 18). 7. Spectrophotometric Determination of Hemiglobin
Hemiglobin (Hi) is the hemoglobin derivative in which the ferrous porphyrin complex is converted to the ferric form by oxidation. This trans-
177
HEMOGLOBIN AND ITS DERIVATIVES
TABLE 18 COMPARISON OF SCO DETERMINED USINGMETHODS 6.1.1
Sample
Method 6.1.1
1 2 3 4 5 6 7 8 9 10
0 100 100 91
AND
6.2.1
Method 6.2.1 ~~
98
0 9 30 47 12
~~
0 102 100 91 96 0 9 27 42 14
formation can take place in different ways. When encountered clinically, it is usually due t o toxic agents. These act either by direct oxidation (NOz-, NOS- after reduction in the intestinal tract to NOz-, Fe+++, etc.), or by indirect oxidation (phenacetin, sulfonamides, salicylazosulfapyridine, aniline and derivatives, nitrobenzene, hydroxylamines, etc.). Besides intoxications, congenital and hereditary hemiglobinemias have been shown t o exist. In many of these cases, hemiglobinemia is the result of a deficiency of the hemiglobin reductase system prevalent in the normal red cell. Another type of congenital hemiglobinemia is due to the presence of an abnormal hemoglobin (HbM), which has an increased sensitivity to oxidation. The stepwise oxidation of the four heme groups to Hi normally goes on continuously in vivo; this process, however, is kept a t an equilibrium by the enzymatic reduction of Hi t o Hb. Due to this equilibrium u p to 0.5% Hi can usually be demonstrated in normal blood. If, for any reason, Hi formation exceeds reduction to Hb, the Hi content of the blood increases, and cyanosis may appear. The determination of hemiglobin in blood is therefore important in clinical medicine. 7.1. PRINCIPLES Hemiglobin can t e estimated in the presence of HbOz, using a twowavelength method (Section 4). A suitable isobestic point is situated at X = 523 nm, while a large difference between E H ~and eHbOI is found a t X = 558 nm. Other possibilities are referred to by Henry (Hl). Another approach to the determination of Hi is to make use of the change in the light absorption spectrum occurring on the addition of CN-. At
178
E. J. VAN KAMPEN AND W. G . ZIJLSTRA
neutral pH, Hi has a maximum at X = 630 nm which disappears on the addition of CN-. Calculations can be made from DS30measured before and after addition of CN-, as has been done by Evelyn and Malloy (El) and Marti (M4). By subsequently converting all hemoglobin present to Hi, using K3Fe(CN)a,the percentage Hi initially present can be determined easily. Although the principle on which these methods are based is attractive, in practice they are disturbed by the occurrence of turbidity. In the modification described in Section 7.3, this difficulty has been overcome by the use of Sterox SE. A third possibility, and an important one from a clinical point of view, is a spectrophotometric screening method. Using this method it is possible to establish within a few minutes whether or not hemiglobin is present in a 0.1-ml blood sample. At the same time it is possible to estimate S H i from the recorded spectrum (Section 7.4). 7.2. DETERMINATION OF SHiMEASURED AT X = 558 1 = 0.01 CM
AND
523
NM;
Blood (2.0 ml) is oxygenated for 5 minutes in a tonometer; 0.5 ml oxygenated blood is drawn into a 2-ml glass syringe containing 0.5 ml 2y0 Sterox SE solution and a metal mixing ring. After thorough mixing and discarding the first three drops, the 0.10-cm cuvette is filled. A 0.09-cm plane parallel glass plate is inserted into the cuvette, leaving a blood layer of 0.01 cm (Fig. 13). The optical density is measured at X = 558 and 523 nm (isobestic wavelength). The Hi concentration is calculated from the equation SHi
(70) = [1.94 - 1.56(D668/D523)]100
(16)
The constants 1.94 and 1.56 have been calculated from a series of 25 measurements of D668/D623 for 0% Hi (100% HbOz) and 100% Hi. OF S HBY ~ CONVERTING HI 7.3. DETERMINATION AND MEASURING AT X = 630 NM
TO
HICN
Blood (0.2 ml) is hemolyzed in a solution containing 4 ml phosphate buffer, 0.1 M (pH 6.8), and 6 ml 1% Sterox SE. The Sorensen buffer contains 34.6 g KHzPOl and 45.24 g NszHPO4.2H20per 250 ml. This stock solution must be diluted 20-fold before use. The hemolysate is divided into two equal parts, designated A and B. A is measured at X = 630 nm; slit width = 0.03 mm; I = 1.000 cm. This measurement yields D1. Next one drop of a 5% KCN solution is added, and after mixing DS30is measured again, yielding Dz. To B one drop of 5y0K3Fe(CN),, solution is added, and after 5 minutes
HEMOGLOBIN AND ITS DERIVATIVES
179
D630is measured. This yields Ds. Next one drop of the 5’% KCN solution is added, and after mixing D630is measured again, yielding D4. All measurements should be made against a blank containing phosphate buffer solution and Sterox SE in the same proportion as present in the sample. The Hi concentration is calculated from the equation SHi
(%) = [ (DI- D.)/(D$ - D4)]1OO
OF S HBY ~ 7.4. ESTIMATION
A
(17)
SCREENING METHOD
Blood (0.1 ml) is mixed with 10 ml of 2% Sterox SE solution. The absorption spectrum of this clear solution is recorded in a Beckman DB or similar spectrophotometer from X = 650 nm down to 560 nm; I = 1.000 cm. The 2% Sterox SE solution is used as a blank.
A (nm)
FIG. 16. T ( % ) / h curves of blood samples containing: -- Hb02; - - - - - - 5% Hi (95% HbOz); -- - 10% Hi (90% HbOz); - - - 15% Hi (85% HbOz); - - - 20% Hi (80% HbOz); the last four down to A = 590 nm.
-
I n Fig. 16, the absorption spectrum from X = 690 down to 590 nm of blood samples containing 5,10,15, and 20% Hi is presented. The spectrum of HbOz down to X = 480 nm has been included. From this figure a n estima~ be made as follows. Suppose that a sample contains aCw t o t . tion of S Hcan quarter millimoles Hi per liter and (1 - u)CHbtot. quarter millimoles HbOz per liter. As Lambert-Beer’s law is obeyed, e = D/cl. We know from = 0.18 and eg: = 3.69. Thus Sections 3.4 and 3.6 that ~g”b”0, 3.69
=
Dg;/(aC,, tot.Z)
(18)
180
E. J. VAN KAMPEN AND W. G . ZIJLSTRA
and
0 * l 8 = Oz3bO,/[(l- U)CHbtot.zl
(19)
Dividing Eq. (18) by Eq. (19) gives
(Dz:/Dg&)[(l - u)/u]
20
or
a/(l - U)
O.O5(DE;/D2&)
And, for small values of a, as a first approximation
SHi(%) = lOOa = 5(DE"/Dg$,)
(20)
For a blood sample containing some Hi, Dm0measured equals the sum of DE; and DEia. A fair approximation of DE:a is obtained by measuring D6I8 of the sample and dividing this value by 45. This factor has been determined experimentally from 10 HbOnabsorption spectra recorded using this method and is valid for this method only. Thus after reading De30and D618 from the recorded spectrum, D E a and 0:: are calculated, and S H ~ is determined using Eq. (20). This screening method proved t o be of great value in clinical emergencies as is demonstrated by the following case report: A 40-year-old fruit sprayer was brought to the hospital one evening with severe cyanosis. During the day he had sprayed an orchard without a mask;
-1
2.000 1,000 0.800
0.600 0.500
(1.300 a200 -
I
0.400
0.100
I
/
>/
'680
'
610
'
6;O
I
5iO
I
510 X;nm]
FAG.17. D/X curves of blood samples taken from a patient with nitrobenzene poison-
ing: - -
- - on admittance to the hospital; -12 hours after treatment,.
181
HEMOGLOBIN AND ITS DERIVATIVES
using nitrobeneene. The absorption spectrum of a blood sample recorded on admission is shown in Fig. 17. From this spectrum the Hi concentration was estimated to be near 20%.A prompt improvement followed intravenous injection of 1 g sodium ascorbate and 1 g methylene blue. The next day an absorption spectrum of a second blood sample was recorded. Hi could no longer be demonstrated (Fig. 17). OF 7.5. COMPARISON
THE
THREEMETHODS
Fresh normal human blood (50 ml) was hemolyzed with 5 ml7% Sterox SE and oxygenated by tonometry for 20 minutes. The total hemoglobin concentration (determined by the method described in Section 2.2.2) was 9.5 quarter millimoles per liter. It can be calculated that the addition of 1.6 mg K3Fe(CN)eto 10 ml of this hemolysate will result in a Hi concentration of 5y0. In this way samples containing approximately 0, 5, 10, 15, and TABLE 19 S HOF ~ 5 BLOODSAMPLES AS DETERMINED BY
THE
THREEMETHODS DESCRIBED
Calculated
Method 7.3
Method 7.2
Method 7.4
0 5 10 15 20
0.1 3 8 13 17
0.5 4 9 14 18
3 8 12 15
20% Hi were made and subsequently analyzed (using the methods described in Sections 7.2, 7.3, and 7.4). The results are given in Table 19. The agreement between the three methods is reasonable. I n a second experiment, a solution containing 100% Hi was made by the addition of excess K3Fe(CN)e t o a hemolyzed blood sample. The excess TABLE 20 S HOF ~ 4 BLOOD SAMPLES AS DETERMINED BY THE METHODS DESCRIBED IN SECTIONS 7.2
AND
7.3
Calculated
Method 7.2
Method 7.3
10 14 15.5 18
10.5 15 18.5
10 14 15 20
182
E. J. VAN KAMPEN AND W. G. ZIJLSTRA
K3Fe(CN)6was then removed by dialysis overnight, and CW tot. was determined (using the method described in Section 2.2.2). Four samples were then made by mixing this solution with a solution containing 100% HbOz with an equivalent total hemoglobin concentration. The S H i was then determined (using the methods described in Sections 7.2 and 7.3). The results, which are in fair agreement, are shown in Table 20. 8. Spectrophotornetric Determination of Sulfhernoglobin
The chemical structure of sulfhemoglobin (SHb), a green hemoglobin derivative, has not yet been completely elucidated. The central iron atom is in the ferrous state. Combination of SHb with CO to SHbCO has been encountered (V2). The structure proposed by Nijveld (N2) for the a-methene group in the porphyrin ring, S -C-
II
breaking the conjugated structure, has not yet been confirmed. As pure SHb solutions have not yet been made, the extinction coefficients of SHb have not been determined accurately. The actual agent in the formation of sulfhemoglobinemia is hydrogen sulfide (HzS) produced in the intestine by the action of bacteria on food residues and absorbed in the large intestine. The HzS normally is excreted from the lungs or destroyed, but if present in excess, or in the presence of substances promoting its action on hemoglobin, SHb is produced (L3). Drugs known to favor the formation of SHb are phenacetin, acetanilide, sulfonamides, salicylazosulfapyridine, nitrites, and nitroglycerin. In some of these cases there seems to be a synergetic action on reductase systems in the intestine leading to the formation of S- from SO4-. In some cases sulfhemoglobinemia has been reported to be accompanied by an increased glutathione concentration in the erythrocytes (Ml). Finally, the occurrence of congenital sulfhemoglobinemia has been described (M6). SHb cannot be converted to active hemoglobin, either in vivo or in vitro. Thus the disappearance of SHb, once formed in the blood, is determined solely by the life span of the erythrocytes concerned. The light absorption spectrum of SHb shows an absorption maximum at X = 618-622 nm (Fig. 18), which can be used for determination purposes. The addition of CN- does not affect this peak, while the Hi maximum at X = 630 nm disappears completely through the formation of HiCN (Section 7.1). Use of this property serves to distinguish between these two hemoglobin derivatives.
HEMOGLOBIN AND ITS DERIVATIVES
-1
183
2.000 1.000
0.800
I
0.600 0.500
0.400
-
-
0.300 I
FIG. 18. D/X curves of sulfhemoglobin (dashed line) and oxyhemoglobiil (solid line). Recorded with a Beckmaii DB spectrophotometer; 2 = 0.005 cm.
T%
0' Ol10 20 -
30 LO
-
50 -
60 -
70 -
520
h(nm)
FIO.19. T%/X curve of a blood sample taken from a patient who took 10 g phenacetin
in 2 days (solid line). The dashed part of the curve denota the shape if only HbOz had been present in the sample.
184
E. J. VAN KAMPEN AND W. G. ZIJLSTRA
8.1. ESTIMATION OF SsHb
BY A SCREENING
METHOD
Blood (0.1 ml) is mixed with 10 ml of 2% Sterox SE solution, and one drop of 5% KCN solution is added. The absorption spectrum of this clear solution is recorded in a Beckman DB or similar spectrophotometer from h = 650 nm down to 560 nm; I = 1.000 cm. The 2% Sterox SE solution is used as a blank. Figure 19 presents the absorption spectrum of a blood sample containing SHb. From this figure an estimation of S S H b can be made as follows. Suppose that a sample contains aCHb tot. quarter millimoles SHb per liter and (1 tot. quarter millholes HbOz per liter. As Lambert-Beer's law is obeyed, E = D/cl. We know from Section 3.4 that EE~o, = 0.27; for &",", we take 13.0 as calculated by E. M. Jope (in L2). Thus 13'0 =
Dggb/(aC,b
t0t.l)
(21)
and 0'27 = Dg&/[(l - a)CHbt&ll
(22)
Dividing Eq. (21) by Eq. (22) gives (D~$b/Dg&)[(l - u)/u] x 50
or a / ( l - a)
0.02(Dggb/Dg&)
and, for small values of a, as a first approximation
For a blood sample containing some SHb, D620measured equals the sum of Dggb and Dg&,. A fair approximation of DE& is obtained by measuring D6'* of the sample and dividing this value by 35. This factor has been determined experimentally from several HbOz absorption spectra recorded using this method and is valid for this method only. Thus after reading DS2Oand D678 from the recorded spectrum, D E a and OLEbare calculated, and SBHb is determined using Eq. (23). The absorption spectrum shown in Fig. 19 was recorded from a blood sample of a patient who had taken 10 g phenacetin in 2 days. Using the method described above, #8Hb was calculated to be approximately 3%. In this case the sample did not contain Hi.
HEMOGLOBIN AND ITS DERIVATIVES
185
REFERENCES B1. Benesch, R., Benesch, R. E., and Macduff, G., Spectra of deoxygenated hemoglobin in the Soret region. Science 144, 68 (1964). B2. Biblhtheca HaematoZ. 18, 110 (1964). B3. Bibliotheca Hmmatol. 21, 215 (1965). C1. Cannan, R. K., Proposal for distribution of a hemoglobin standard. Science 122, 59 (1955). C2. Cannan, R. K., Hemoglobin standard. Science 127, 1376 (1958). C3. Chilcote, M. E., and O’Dea, A. E., Lyophilized carbonylhemoglobinas a colorimetric hemoglobin standard. J . BWZ. Chem. 200, 117 (1953). Dl. Drabkin, D. L., Spectroscopy, photometry and spectrophotometry. I n “Medical Physics” (0. Glaser, ed.), Vol. 11, pp. 1039-1088. Year Book Publ., Chicago, Illinois, 1950. D2. Drabkin, D. L., Metabolism of the hemin chromoproteins. Physwl. Rev. 31, 345 (1951). D3. Drabkin, D. L., and Austin, J. H., Spectrophotometric studies. I. Bpectrophotometric constants for common hemoglobin derivatives in human, dog and rabbit blood. J . Bwl. Chem. 98,719 (1932). D4. Drabkin, D. L., and Austin, J. H., Spectrophotometric studies. 11. Preparations from washed blood cells; nitric oxide hemoglobin and sulfhemoglobin. J . BWZ. Chem. 112,51 (1935). D5. Drabkin, D. L., and Austin, J. H., Spectrophotometric studies. V. A technique for the analysis of undiluted blood and concentrated hemoglobin solutions. J . Biol. Chem. 112, 105 (1935). El. Evelyn, K. A,, and Malloy, H. T., Micro determination of oxyhemoglobin, methemoglobin and sulfhemoglobin in a single sample of blood. J . BioZ. Chem. 126, 655 (1938). F1. Falholt, W., Blood oxygen saturation determined spectrophotometrically. Scand. J . Clin. Lab. Invest. 16, 67 (1963). G1. Geubelle, F., Sbmi-micro methode spectrophotombtrique de dbtermination du rapport oxyhbmoglobine/hbmoglobine dans le sang. Clin. Chim. A d a 1, 225 (1956). G2. Gordy, E., and Drabkin, D. L., Spectrophotometricstudies. XVI. Determination of the oxygen saturation of blood by a simplified technique, applicable to standard equipment. J . Bwl. Chem. 227, 285 (1957). G3. Green, P., and Teal, C. F. J., Modification of the cyanmethemoglobin reagent for analysis of hemoglobin in order to avoid precipitation of globulii. A m . J . CZin. PathoZ. 32, 216 (1959). H1. Henry, R. J., “Clinical Chemistry, Principles and Technics,” p. 749. Harper, New York, 1964. H2. Holling, H. E., MacDonald, I., O’Halloran, J. A., and Venner, A., Reliability of a spectrophotometric method of estimating blood oxygen. J . AppZ. PhysioZ. 8, 249 (1955). J1. JonxiS, J. H. P., and Boeve, J. H. W., A spectrophotometric determination of oxygen saturation in small amounts of blood. Acta Med. Scand. 166, 157 (1956). K1. Klungsrayr, L., and Stoa, K. F., Spectrophotometric determination of hemoglobin oxygen saturation. Rec. Trav. Chtm. 74, 571 (1955). L1. Lambertsen, C. J., Bunce, G. H., Drabkin, D. L., and Schmidt, C. F., Relationship
186
E. J. VAN KAMPEN AND W. G. ZIJLSTRA
of oxygen tension to hemoglobin oxygen saturation in the arterial blood of normal men. J . Appl. Physwl. 4, 873 (1952). L2. Lemberg, R., and Legge, J. W. “Hematin Compounds and Bile Pigments,” p. 491. Wiley (Interscience), New York, 1949. L3. Lemberg, R., and Legge, J. W. “Hematin Compounds and Bile Pigments,” p. 523. Wiley (Interscience), New York, 1949. M1. McCutchem, A. D., Sulfhaemoglobinaemiaand glutathione. Lancet 11,240 (1960). M2. Malenstrom, G., and Sjostrand, T., Physiological variations in the endogenous formation of carbon monoxide. Acta Physwl. Scand. 27, 231 (1952). M3. Mann, J. D., Woodson, M. A., Hoffman, R. G., and Martinek, R. G., The relation between reported values for hemoglobin and the transfusion rate in a general hospital. Am. J . Clin. Pathol. 32, 225 (1959). M4. Marti, H. R., “Normale und anormale menschliche Hamoglobine,” p. 49. Springer, Berlin, 1963. M5. Mason, E. C., and Adarraga-Elizaran, A., Standardization of hemoglobin solutions by iron determination, J . Clin. Pathol. 16, 604 (1963). M6. Miller, A. A., Congenital sulfhaemoglobinaemia.J. Pediat. 61, 233 (1957). M7. Minkowski, A., and Swierczewski, E., The oxygen capacity of human foetal blood. I n “Oxygen Supply to the Human Foetus” (J. Walker and A. C. Turnbull, eds.), pp. 237-263. Blackwell, Oxford, 1959. N1. N a b , G. G., Spectrophotometricdetermination of hemoglobin and oxyhemoglobin in whole hemolyzed blood. Science 113, 723 (1951). N2. Nijveld, H. A. W., Properties and structure of sulfhemoglobin. Rec. Trav. Chim. 62, 293 (1943). P1. Paul, K. G., and Thorell, H., A colorimetrical carbon-monoxide-hemoglobin method of determination for clinical use. Acta Physiol. Scand. 4, 285 (1942). R1. Refsum, H. E., Spectrophotometric determination of hemoglobin oxygen saturation in hemolyzed whole blood by means of various wavelength combinations. Scand. J . Clin. Lab. Invest. 9, 190 (1957). R2. Refsum, H. E., and Sveinsson, S.L., Spectrophotometric determination of hemoglobin oxygen saturation in hemolyzed whole blood. Scand. J . Clin. Lab. Invest. 8, 67 (1956). R3. Remmer, H., Die Standardisierung des roten Blutfarbstoffes durch Hamiglobincyanid. 11. Mitteilung. Eisengehalt und OzBindungsvermogen von menschlichem Blut. Aich. Exptl. Pathol. Pharmakol. 229, 450 (1956). S1. Salvati, A. M., Tentori, L., and Vivaldi, G., The extinction coefficient of human hemiglobincyanide. Clin. Chim. Acta 11, 477 (1965). S2. Sjostrand, T., The formation of carbon monoxide by in vitro decomposition of haemoglobin in bile pigments. Acta Physwl. Scand. 27, 231 (1952). V1. van Kampen, E. J., and Klouwen, H., Spectrophotometric determination of carboxyhemoglobm. Rec. Trav. Chim. 73, 119 (1954). V2. van Kampen, E. J., Volger, H. V., and Zijlstra, W. G., Spectrophotometric determination of carboxyhemoglobinin human blood and the application of this method to the estimation of carbonmonoxide in gaamixtures. Koninkl. Ned. Akad. Wetenschap. Proc. CS7, 320 (1954). V3. van Kampen, E. J., and Zijlstra, W. G., Standardization of hemoglobinometry. 11. The hemiglobincyanidemethod. Clin. Chim. Acta 6 , 538 (1961). V4. van Oudheusden, A. P. M., van de Heuvel, J. M., van Stekelenburg, G. J., Siertsema, L. H., and Wadman, S.K., De ijking van de hemoglobinebepaling op basis
187
HEMOGLOBIN AND ITS DERIVATIVES
Z1.
22. 23. 24.
van ijser (Calibration of the haemoglobin determination on the basis of iron). Ned. Tijdschr. Geneesk. 108, 265 (1964). Zijhtra, W. G., and van Kampen, E. J., Standardization of hemoglobinometry. 1. The extinction coefficient of hemiglobincyanide a t h = 540 mp; Clin. Chim. Acta 6, 719 (1960). Zijhtra, W. G., and van Kampen, E. J., Standardization of hemoglobinometry. 111. Preparation and use of a stable hemiglobincyanide standard. Clin. Chim. Acta 7 , 96 (1962). Zijlstra, W. G., and Muller, C. J., Spectrophotometry of solutions containing three components, with special reference to the simultaneous determination of carboxyhemoglobin and methemoglobm in human blood. Clin. Chim. Ada 2, 237 (1957). Zondag, H. A., and van Kampen, E. J., Determination of copro- and uroporphyrin in urine. Clin. Chim. Acta 1, 127 (1956).
eZ&..
This Page Intentionally Left Blank
BLOOD- COAGULATlON FACT0R VI II: GENETICS, PHYSIOLOG1CAL CONTRO1, AN D BIOASSAY
G. 1. C. lngram Department of Haematology, Louis Jenner Laboratories, St. Thomas’ Hospital and Medical School, London, England
1. Introduction.. . . . . . . . . . . . . . . . . . . . . . . . . . . . References, Section 1 ................................................. 2. Genetics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Synopsis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. The Growth of Data.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. Genetic Models.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.4. Antibody Studies.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.5. The Factor VIII Locus on the X-Chromosome ........... References, Section2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3. Physiological Control. ................... ........... 3.1. Synopsis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Long-Term Changes.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Short-Term Changes.. .. ...................... 3.4. Investigations into the N e in Factor VIII Concentration . . . . . . . . . . . . . . . . ................. 3.5. Possible Mechanisms for or VIII.. . . . . . . . . . References, Section 3 . ..............................
....................................................
4.1. Synopsis.. ............. 4.2. Some Principles of the Bioassay of 4.3. Sources of Error in Clotting-Factor 4.4. Calculating the Potency Ratio.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References, Section 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
189 190 191 191 191 194 201 204 205 207 207 207 209 211 216 218 222 222 223 225 226 235
1. Introduction
The numeral VIII has been agreed upon internationally to designate the clotting factor which is missing from the blood of hemophiliacs. Other names include antihemophilic globulin (AHG), in common British usage, and antihemophilic factor (AHF), more commonly used in the United States; it is also known as antihemophilic factor A, because the bleeding disorder now properly called “hemophilia” is sometimes distinguished from allied conditions as hemophilia A. Factor VIII is believed to be an az-globulin (Bl) and to have a high molecular weight (Wl). In the sequence of reactions taking place when blood clots, the “resting” plasma form of factor VIII is thought to be acted upon by “activated” factor IX (Ml) and, when 189
190
G. 1. C. INGRAM
thus “activated,” reacts with factor X. Phospholipid probably also participates in the “activation” of factor VIII. The meaning of “activation” in terms of structural molecular changes has been discussed by Vroman (Vl). The rate of growth in our knowledge of blood coagulation is becoming increasingly rapid, and it is a sign of the times that the section of this volume devoted to clotting can be wholly occupied by a review of three virtually distinct fields of work related to a single clotting factor. Indeed, the aspect of factor VIII perhaps receiving most attention a t the time of writing is its role in the coagulation sequence; but, during the present intensive phase of this work, a review would become so quickly out of date that it has seemed better to concentrate on fields where research is momentarily less active. While factor VIII has been chosen by the present reviewer because it is his main interest, other writers, even if less prolix, could have filled as many pages with a corresponding discussion of almost any other clotting factor. Factor VIII provides, however, an appropriate subject for the “Clinical Chemistry” series because the work to be described springs largely from clinical observations on hemophiliacs; and although the techniques employed are seldom those of classical chemistry, they have provided much basic knowledge of the biological activities of the factor VIII molecule. Only in recent times have investigators of blood coagulation felt able to refer to “concentrations” of plasma coagulation factors. They have long retained a deep-seated suspicion that many of the factors described would eventually prove not to be distinct chemical substances but only “activities,” or even more ephemeral experimental phenomena (as indeed happened with factor VI, now omitted from the international list). It seems fair now to accept factor VIII as a real entity, however, even though its recognition still depends entirely on biological activity and is at times uncertain (see Section 3.4, below). Although something of the old caveat is therefore still justifiable, the term “concentration” will be freely used for factor VIII in this review.
REFERENCES, SECTION1 B1. Barkhan, P., Lai, M., and Stevenson, M., Antihaemophilic factor (factor VIII): 2 Brit. J . Haematol. 9, 499-505 (1963). an ~ r globulin. M1. Macfarlane, R. G., Biggs, R., Ash, B. J., and Denson, K. W. E., The interaction of factors VIII and IX. Brit. J . Haematol. 10, 530-541 (1964). V1. Vroman, L., Effects of hydrophobic surfaces upon blood coagulation. Thromb. Diath. Haemorrhag. 10, 455-493 (1964). W1. Wagner, R. H., and Thelin, M., Antihemophilic factor: current status of purification, chemical and physical characterization. In “Hemophilia and Hemophilioid
BLOOD-COAGULATION FACTOR VIII
191
Diseases” (K. M. Brinkhous, ed.), pp. 3-10. Univ. of North Carolina Press, Chapel Hill, North Carolina, 1957.
2. Genetics
2.1. SYNOPSIS Classical studies of the recessive, sex-linked disorder hemophilia provided evidence that a gene concerned with the synthesis of factor VIII must be situated on the X-chromosome. When it was discovered that a reduction in factor VIII was commonly present also in von Willebrand’s disease, with a somatic dominant inheritance, it became clear that another gene, on an autosome, must also be involved. Cross-transfusion experiments between patients suffering from hemophilia and von Willebrand’s disease showed that hemophilic blood would stimulate factor VIII synthesis in von Willebrand’s disease, but not vice versa. These data have made it possible to construct a number of alternative genetic models now being submitted to critical experimentation. Some progress has also been made in mapping the position of the factor VIII locus on the X-chromosome. OF DATA 2.2. THEGROWTH
2.2.1. Hemophilia and, von Willebrand’s Disease Otto (01)of Philadelphia, credited with the first clear account of hemophilia (B5), described a lifelong bleeding disorder affecting males and transmitted by unaffected females. The classical studies of Bulloch and Fildes (B7) definitively associated the name “hemophilia” with this pattern of inheritance, called sex-linked recessive, and were later confirmed in hemophilic dogs by obtaining the expected progenies from the various possible crosses of affected and unaffected males with normal and carrier females (B6, F1, H2). Thirteen years ago Aggeler et al. (A3) and Biggs et al. (B3) distinguished a closely allied disorder, plasma thromboplastic component (PTC) deficiency or Christmas disease, having a similar mode of inheritance but due to deficiency of a different clotting factor (factor IX). It was thus apparent that a t least two genes on the normal X-chromosome must be concerned in the synthesis of clotting factors, one being factor VIII, but little development of genetic theory could be undertaken with only these facts available. Another lifelong, hereditary bleeding disorder had been described by von Willebrand (Vl, V2) among the inhabitants of the Aaland Islands, in the mouth of the Gulf of Bothnia. This condition affects both sexes but is
192
G . I. C. INGRAM
transmitted only by affected persons; the bleeding symptoms also differ somewhat from those of hemophilia. The disorder is further distinguished by a prolonged bleeding time but a normal or near-normal whole-blood clotting time, whereas characteristically in hemophilia the bleeding time is normal, but the clotting time greatly prolonged, a t least in the severe, so-called classical hemophilia, as Bulloch and Fildes (B7) had realized. It was therefore a quite unexpected discovery (A4, 52, L1, N2) that the plasma concentration of factor VIII might be distinctly reduced in cases of von Willebrand’s disease, although not usually to the very low levels encountered in severe hemophilia. Thus it became clear that one of the autosomes (the somatic chromosomes) must also carry a gene concerned in factor VIII synthesis, because of the different mode of inheritance of this disease. A small series of cases implying yet a third gene controlling factor VIII synthesis has been described in recent years. In these cases factors V and VIII are reduced together and the defect seems to be transmitted as an autosomal recessive disorder (Jl). It also appears from studies of normal populations that the factor VIII concentration of an individual is correlated with the concentrations of both his parents; this has been taken to imply the effects of (several?) autosomal genes (Pl). Furthermore, persons with blood group A tend to have higher concentrations than group 0 subjects (P2). These findings, although suggesting that more than one autosomal gene is involved, are not yet sufficiently clear to be usefully included in genetic models, and will not be further discussed here.
2.2.2. The Bleeding-Time Factor The long bleeding time in von Willebrand’s disease was mentioned above. This was at first thought to be due to an abnormality of the platelets (V3), but was not confirmed, and so a capillary abnormality was postulated and became generally accepted, some workers feeling confident that morphological abnormalities of the small vessels could be observed (B4). It was thus surprising to find (N3) that an infusion of normal plasma would temporarily shorten the bleeding time in von Willebrand’s disease, a phenomenon now amply confirmed. Although the prolonged bleeding time is distinct from the deficiency of factor VIII, it is mentioned here because it has been natural to attempt to account for both features of the disorder when constructing genetic models, and will be referred to below. In fact, the principle in normal (or hemophilic) plasma having the ability to shorten the bleeding time in von Willebrand’s disease is known as the %on Willebrand factor” or “bleeding-time factor” (Bl), and is now known to be a necessary component in the mechanism whereby adenosine diphosphate
BLOOD-COAGULATION FACTOR VIII
193
causes the clumping of platelets (Hl, Sl), and presumably their adhesion to the edges of a rent in a damaged blood vessel. 2.2.3. Stimulated Synthesis of Factor V I I I in von Willebrand’s Disease To return to the concentration of factor VIII in von Willebrand’s disease: at about the time when the corrective value of normal plasma on the bleeding time was discovered, it was also found that an infusion of a fraction of nonnal plasma led to a prolonged rise in the factor VIII concentration in the patient’s blood (N2). In contrast, the effect of infusing normal plasma in classical hemophilia, was already familiar; it was known that after a single infusion the rise in factor VIII in the patient’s plasma bore a reasonably predictable relationship to the patient’s plasma volume and to the factor VIII content of the infused material, and that the concentration then immediately began to fall at more than 60% per day. The effect in von Willebrand’s disease was strikingly different; the initial rise continued far beyond that anticipated from the factor VIII content of the plasma infused, and the high concentration was maintained for about a day, eventually returning to the original low level some 3 M 8 hours from the time of the infusion. This seemed to mean that, following the infusion, the patient had been enabled to synthesize a quantity of factor VIII considerably in excess of the amount infused, assuming that the rate of disappearance of factor VIII was the same in von Willebrand’s disease as in hemophilia. It was even more striking to find that a similar response could be obtained by infusing hemophilic plasma (Cl, C2, N l ) , without detectable factor VIII activity, for which no explanation other than “new synthesis” seemed possible. In contrast, infusing von Willebrand plasma into hemophiliacs produced no more rise in the patient’s factor VIII than would be expected from the factor VIII content of the infusate (Bl, Cl). The most economical hypothesis at first was to suppose that the bleeding-time factor and the component stimulating factor VIII synthesis were the same substance, but it was shown at more than one center that a certain factor VIII concentrate made in Great Britain could stimulate synthesis in von Willebrand’s disease, but did not affect the bleeding time. Also, Egeberg (El) showed that serum prepared from an exercised donor could shorten the bleeding time in a case of von Willebrand’s disease without raising the factor VIII concentration [although a factor VIII response following serum infusions has been reported by Biggs and Matthews (B2)J. Therefore it now seems necessary to consider the possibility of two separate components. A similar “new synthesis’’ of factor VIII has been observed in von Willebrand’s disease in pigs following the infusion of plasma and serum from normal pigs (M4).
194
G. I. C. INGRAM
2.2.4. Summary of Findings The main facts now available on which to build a genetic hypothesis are as follows: first, that factor VIII synthesis requires the activity of a t least two distinct genes, one on the X-chromosome and the other on an autosome; and that mutation of the X-chromosome gene might lead t o complete failure to synthesize any biologically active product (the situation in classical hemophilia, with zero factor VIII activity in the plasma), but that mutation of one autosome gene usually leads to only partial reduction (note the implication that cases of von Willebrand’s disease are assumed to be heterozygotes, because of the apparent rarity of the disorder in most populations); second, that cross-transfusion experiments show a component to be present in normal and in hemophilic blood which can lead to rapid synthesis of factor VIII in von Willebrand’s disease, but that von Willebrand blood has no such effect in hemophilia. This implies that factor VIII is reduced in the plasma in von Willebrand’s disease because there is failure adequately to synthesize some component being made normally in hemophilia. 2.3. GENETICMODELS In the foregoing discussion of hemophilia, the emphasis has been on the “classical1’disorder, in which no factor VIII can be detected in the patient’s plasma. There are, however, many cases of hemophilia in which the symptoms are relatively mild (the patients being free of the well-known spontaneous bleeding and suffering only from abnormally prolonged posttraumatic bleeding) and in which the plasma factor VIII concentration is less seriously reduced. It seems reasonable to postulate that many different mutations may occur in the sex-linked (X-chromosome) gene, and that in the milder cases the gene product retains some biological activity, whereas in the classical cases the product either cannot be formed a t all or is functionless in the clotting system. Therefore, no special difficulty exists in dealingwith variations in the severity of hemophilia, where the hemophiliac, beinga male, has only one X-chromosome. It is thus convenient to construct models in terms of classical hemophilia with total lack of factor VIII activity in the plasma, and then to apply them to mild cases with the appropriate modification. With this understanding, the remainder of the discussion will be in terms of the severe disorder. The position of the heterozygous (carrier) female can now be considered. She has the same complement of one normal X-chromosome as the normal male., which appears to account for het normal phenotype. A complication arises if the Lyon hypothesis (L3) is considered to apply: this states that only half the female complement of X-chromosomes are functional, so that
BLOOD-COAGULATION FACTOR VIII
195
the female carrier would operate only half the number of normal X-chromosomes active in the normal rnale. Compensatory mechanisms apart, this leads to the expectation of factor VIII concentrations in heteroaygous women (carriers of the severe disorder) of around half those in normal women, which a number of authors have in fact reported as an average finding (M3, P3) although many individual heterozygotes have a normal concentration. (A woman must be heteroaygous for hemophilia if she is the daughter of a hemophiliac and her mother is homoaygous normal; also, she may be presumed to be heteroaygous if she is the mother of more than one hemophilic son, or has a hemophilic grandson by more than one daughter, or one of each, or has one hemophilic son or grandson and also a hemophilic brother or maternal uncle or a hemophilic cousin who is the son of a maternal aunt.) Even if heteroaygous women normally have factor VIII concentrations of about half that of normal females, their lack of symptoms is not difficult to accept, because the normal range in the population lies between about half and twice the mean factor VIII concentration (P2), and hemostasis is virtually normal even if the plasma concentration is as low as 20-30% of the mean. Carriers of the milder forms of the disorder would be correspondingly even less likely to have symptoms, although some such carriers have been found with considerable expression of the abnormality (G3).
2.3.1. A Simple Model On the lines just discussed, it is possible t o construct a simple model of the genetics of factor VIII synthesis in von Willebrand’s diseaseand hemophilia. This model will be discarded because it will not explain all the facts, but it may be easier to accept the greater complexities of models proposed by J. B. Graham and outlined below when the defects of a simpler model have been grasped. The simplest explanation of the results of cross-transfusion experiments seems to be to suppose that factor VIII is synthesized in two stages: that the “von-Willebrand” genes normally create an intermediate product which the “hemophilia” gene then converts to the final product, factor VIII. This may be written
von Willebrand genes on paired autosomes
hemophilia gene on X-chromosome
196
G . I. C. INGRAM
I
von Willebrand genes
(mutant hemophilia gene)
and that in von Willebrand’s disease normal von Willebrand gene
(mutant von Willebrand gene)
hemophilia gene
The model would explain the results of cross-transfusion experiments by supposing that hemophilic blood contains a normal complement of intermediate product, which could be converted to factor VIII after infusion into the von Willebrand’s patient. This model, however, has two defects. First, it assumes a serial process of factor VIII synthesis, which is not how proteins are currently thought to be formed (D2). Second, it predicts a factor VIII concentration of at least 50% in von Willebrand’s disease. The illustration shows a mutant yielding no active intermediate; with a mutant synthesizing a product with some activity, more than 5Oy0 of factor VIII would be predicted. I n fact, the factor VIII concentration may be much lower than 5oa/, in von Willebrand’s disease, and to explain this on the suggested model it would be necessary to postulate, for instance, that the inactive product of the mutant could variably block the handling of its normal counterpart by the product of the hemophilia gene, and this might presuppose the hemophilia gene product to have a greater affinity for the abnormal than the normal intermediate, or that more abnormal than normal intermediate was formed. Furthermore, the model does not explain how a mutation leading to a diminished output of h a 1 product can be dominantly inherited, as is von Willebrand’s disease. 2.3.2. The “Regulatory” Model To avoid these dificulties, J. B. Graham (G2, G4)proposed the model illustrated in Fig. 1. Analogies can be found in protein synthesis in other
BLOOD-COAQULATION FACTOR VIII
--
197
Autosome
+
t
t
t
+ I
AH F c-
X- chromosome
FIG. 1. Graham’s (G4) “regulatory” hypothesis for normal synthesis of factor VIII: indicates the normal or “wild-type” gene constitution. Rv.~.d.+ and O+ SG: . . . SG: are the autosomal regulators and operons (operator, with structural genes 1 through n) concerned in von Willebrand’s disease. RAHF+ and OmF+SGAHF+ are the regulator and operon concerned in hemophilia;while the A.H.F. operon must be on the X-chromosome, the regulator could be on any chromosome, and could thus be single if on the X-chromosome, or paired if autosomal. Regulators are conceived of as acting by repression, so that full operation of the R genes would result in total suppression of the output of the corresponding structural genes, and a deficit of repressor would lead t o more than the normal output of final product. The crux of this model is the concept that the product of the autosomal genes (von Willebrand operons), marked “effector,” blocks the activity of RmF+ (indicated by the heavy black line and by the expression R E + R‘ E‘), and thus permits activity in OAHF+SCfLHF+,and thus synthesis of factor VIII (AHF). (V.F.?) is intended to suggest that the effector might also be the von Willebrand factor or bleeding-time factor; this feature of the model was proposed before it became apparent that factor VIII synthesis and the shortening of the bleeding time are mediated through distinct components. Note that the repressor product of each R v . w . d . + acts indifferently on each target operator gene, and that the effector substance produced by each structural gene series is indistinguishable in its ability to block the repressor activity of RAHF+. +
+
+
organisms although none is yet known in man (E2). The particular virtue of this model is that it allows for any degree of factor VIII deficiency in either hemophilia or von Willebrand’s disease. Hemophilia can be represented as before by the reasonable hypothesis of a family of possible mutations in the hemophilia operon on the X-chromosome, some yielding a product with reduced activity, others either no product or a product which is biologically quite inactive. The state of affairs in von Willebrand’s disease is illustrated in Fig. 2, where the von Willebrand mutation is located
198
0. I. C. INGRAM
Autosome
i
AH F Crr.
X-chromosome
FIG. 2. Graham’s (G4) “regulatory” hypothesis for heterozygous von Willebrand’a and the corresponding normal gene R v . ~ . d . + disease. The mutant gene is shown as Rv.w.d. as before. The mutation of von Willebrand’s d i s e q is thus allocated to the regulator. It is then necessary to postulate that the abnormal product is a “superrepressor” (indicated by the heavy black line), that is, that it has a greater repressor effect upon its target operon than the normal product. This means that the effector product of the operon is less than normal (indicated by the word “effector” being written more lightly than in Fig. l),so that it in turn has less than the normal inhibitory effect on the hemophilia-gene repressor (indicated by the long dotted line replacing the heavy black line of Fig. 1).Thus the X-chromosome operon is excessively repressed and less factor VIII than normal is produced, although what is produced is of normal structure.
in the regulator rather than the structural gene group. The detailed consequences of this proposal are explained in the legend to the figure. Any degree of factor VIII deficiency can be produced by imagining a family of mutations in the von Willebrand regulator, producing a series of superrepressors of varying degrees of activity, causing interference of varying degrees in the output of effector. Furthermore, this model explains how a gene product (i.e., the effector) less active than the normal can be inherited as a “dominant” characteristic, if its normal function is to block a repressor. According to this model, the infusion of hemophilic plasma into a patient with von Willebrand’s disease would increase the plasma concentration of effector, which in turn would block the action of the hemophilia-gene repressor toward a normal level of activity, and thus increase the synthesis of factor VIII by the X-chromosome operon. A crucial implication is that the transfusion effect should operate in the same way whether the von Willebrand patient is heterozygous or homozygous, for in either case his (or her) normal X-chromosome operon should synthesize normal factor
199
BLOOD-COAGULATION FACTOR VIII
VIII when the hemophilia-gene repressor is blocked by the infused effector. Another phenomenon explicable by this model is discussed below (Section 3.5.2). 2.3.3. The “Combining Subunit” Model A critical test of the “regulatory” hypothesis would therefore be to transfuse a patient homozygous for von Willebrand’s disease with hemophilic plasma and see whether or not his factor VIII level rose. If it failed to rise in the same manner as in heterozygous von Willebrand’s disease, the model would be seriously called in question. The problem is to identify the homozygotes, who might be expected among the affected families in the Aaland Islands but might otherwise be extremely rare. Graham (B8), working in North Carolina, however, has located one kindred in which two severely affected members may be homozygous for von Willebrand’s disease. They are double third cousins, one male and one female, each the offspring of a second-cousin marriage, such that four of their eight grandparents are themselves grandchildren of one marriage. In one case, that of the girl, it is likely that both parents are mildly affected. The hematological findings in the parents and daughter are shown in Table 1. The striking features are the expected “new synthesis” of factor VIII TABLE 1 “M” KINDRED5’b
GR-4HAM’S
Bleeding time (Ivy’s method) (min) Subject Daughter Mother Father Normal range
>15
Plasma factor VIII concentration (% of average normal) Before infusion
After infusion
6
50-200
-
From Barrow el al. (B8). Bleeding time, and plasma factor VIII concentrations, before and after infusion of hemophilic plasma, in subjects V/1, IV/l, and IV/2 in the “M” kindred. 5
in the parents on infusion of hemophilic plasma, and the very low response in the daughter. A similar effect was observed when the experiment was repeated on the severely affected boy, although it is less clear on the laboratory evidence that his parents are both mildly affected, so his status as a homozygote is more doubtful. This evidence, however, has seriously questioned the validity of the “regulatory” model. Graham’s “combining
U
von Willebrand‘s Disease Normal male Loci: Genes:
Autosomal
+/+
Product: Combination of Aand B-chains:
Free A-chains:
1
1
A+
A+
+
X
+
1
Bi
A+ B+ = 100% factor VIII activity A+ in ex-
Hemophilia (male)
+/+ 1
1
X h
A+
A+
Bh
Autosomal
I
Heterozygous male Autosomal + / v
1
1
Af
Av
+
A+ Bh = 0% factor VIII activity A+ in ex-
Homozygous male
X
Autosomal v / v
+ 1
1
Av
B+
1
A’
X
+ I
B+
+
50% factor Av B+ = O%factor VIII activity VIII activity
=
} in excess
Av in excess
tA+
+AV
Ro.3. Graham’s (G4) “Combining S,ubunit” hypothesis for synthesis of factor VIII in the normal subject, in hemophilia, and in von Willebrand’s disease: + indicates the normal or “wild-type” gene constitution; and I,indicate the von Willebrand and hemophilic mutants, respectively. (For comparison with hemophilia, only male subjects are represented.) It is supposed that each autosomal locus (t.he von Willebrand genes) and the X-chromosome locus (the hemophilia gene) create products at the same rate, molecule for molecule, and that these are polypeptide chains (designated A and B, respectively), which subsequently combine (in the cytoplasm of the synthesizing cells?) to form factor VIII. Since two A-chains are formed for each B-chain, an excess of free A-chains remains. In hemophilia all the B-chains are abnormal (Bh), and thus the completed molecules (A+ Bh) are all abnormal, despite the normal A+ component; the excew consists entirely of normal A-chains (A+). In von Willebrand‘s disease some (heterozygote) or all (homozygote) of the completed molecules are abnormal because they contain abnormal A-chains (Av). The excess will contain free Av-chains. Note that by this model, abnormal factor VIII is predicted in von Willebrand’s disease (Av B+) aa well as in hemophilia (A+ Bh).
+
+
Q I4
+
fg
BLOOD-COAGULATION FACTOR VIII
20 1
subunit” model therefore makes an attractive alternative (Fig. 3). By this hypothesis, the infusion of hemophilic plasma into a case of von Willebrand’s disease introduces a quantity of normal A-chains, which then enter cells (because the same effect cannot be obtained by mixing plasmas in vitro) and there combine with the normal B-chains to form normal factor VIII. It must probably be supposed that the B-chains have a greater aanity for the normal introduced chains than for the endogenous abnormal chains, t o account for the very considerable synthesis of factor VIII often observed after infusions. This creates difficulty in explaining the great variation in factor VIII concentrations observed in von Willebrand’s disease, particularly the low levels, because this model, like the first considered (Section 2.3.1), predicts 50%. An increased affinity of B-chains for abnormal A-chains (McLester and Graham, Mla) cannot be invoked, for the opposite relationship has just been posited. One possible explanation is that more subjects are homozygous for von Willebrand’s disease than is supposed, and that the spectrum of factor VIII concentrations encountered is due to the existence of a family of possible mutations in the autosomal loci which may thus give rise to A-chains with various degrees of partial activity. Barrow and Graham (Bl) have pointed out that von Willebrand’s disease shows great variability of expression of the mutant gene, and have published a kindred of some 60 members illustrating this. Therefore it is possible that family studies have failed to detect the inheritance of the condition from both parents in a number of homoaygotes. 2.4. ANTIBODYSTUDIES
It has been mentioned that zero factor VIII activity in the plasma might be due either to no product or to a biologically inert molecule being produced. These alternatives are of course often met in studying the products of mutant genes, and it may be possible to choose between them by applying an immunological test. This depends on the concept that an abnormal product, devoid of its ordinary biological functions, may yet be sufficiently normal in structure to react (“cross-react”) with an antibody to the normal product. If these maneuvers are followed and neutralization of the antibody can be demonstrated, this is taken to indicate the presence of ‘(cross-reactingmaterial” (CRM), and the extract tested is said to be CRM-positive. If neutralization of the antibody does not occur, the extract is said to be CRM-negative, and it is concluded that the mutant gene forms a product so abnormal as to lack even the antigenic sites, or else no product whatever. To apply this test satisfactorily a sufficiently specific antiserum is necessary. In dealing with factor VIII it is difficult to obtain a pure sample of
202
G . 1. C. INGRAM
antigen with which to prepare the antiserum, for this factor is identifiable only by its biological activity. While the factor can be concentrated several thousandfold, there is always the possibility that inert but antigenic material may be still present, which would confuse the results of the experiment. However, factor VIII-deficient patients have been encountered from time to time whose blood has possessed the ability to destroy added factor VIII from normal plasma. Although fortunately not very common (these patients present a serious clinical problem because the hemostatic mechanism fails to respond to infusions of material containing factor VIII), the condition is now well recognized, and extensive tests in a number of such cases have shown that the inactivating material reacts with no other known clotting factor. Therefore it seems reasonable to regard the abnormal constituent as a specific anti-VIII antibody, despite a frequent failure to demonstrate convincing precipitin lines in agar diffusion. Two such cases, St. and De., were studied by Goudemand et al. (Gl), and in these instances rather indefinite precipitin lines were obtained on Ouchterlony plates. The authors then attempted to use the serum of patient De. for CRM testing of plasma and serum from other patients. The procedure was immensely complicated, for the only specific test of antibody neutralization was the loss of ability to neutralize normal factor VIII, and this involved the following steps: first, incubating De.’s serum with the material to be tested; second, adding a fixed quantity of normal plasma (as a source of normal factor VIII) and continuing the incubation; and, finally, adding this mixture to a blood-coagulation test to assay the residual factor VIII remaining in the normal plasma added at the second step. It was necessary not only to adjust the quantities a t each step so as to achieve a final factor VIII concentration appropriate to the dose-response curve of the assay system, but also to include sufficient blank runs adequately to control the effect on the assay system of the numerous other components present in the mixture of test material and De.’s serum. While paying tribute to the ingenuity and patience of the authors, a certain reserve is inevitable in considering their findings. Some relatively clear-cut results are shown in Table 2. The data are interpreted as showing complete neutralization of the anti-VIII antibody by normal serum, partial neutralization by hemophilic plasma, and none by von Willebrand plasma. From this the authors deduce that both normal serum and hemophilic plasma contain factor VIII in an inactive form, but that no such material is present in von Willebrand’s disease. The work of Goudemand et al. (Gl) received some support from the studies of Adelson et al. (A2), who prepared in rabbits an anti-factor VIII
203
BLOOD-COAGULATION FACTOR VIII
TABLE 2 NEUTRALIZATION TESTSOF ANTI-FACTOR-VIIIBY HEMOPHILIC AND VON WILLEBRAND PLASMAP~ Mean residual anti-factor-VIII (units)
Test material Normal serum (29 testa) Hemophilic plasma (13 tests) von Willebrand plasma (14 tests)
Buffer control (full residual activity)
Normal plasma control (‘‘maximum consumption” of antibody activity)
Test material
3.38
1.76
1.82
3.21
1.35
2.14
3.45
1.70
3.33
From Goudemand et al. (Gl).
* The data are interpreted as showing complete neutralization
of antibody by normal serum, partial neutralization by hemophilic plasma, but no neutralization by von Willebrand plasma.
antiserum and used it to follow the disappearance of labeled factor VIII infused into normal human recipients, and obtained a weighted mean half-survival time of 2.9 days in five subjects. This is considerably greater than the numerous values, all below 20 hours (Al), which have been reported by various observers for the rate of disappearance of factor VIII infused into hemophilic subjects, as measured by biological assays. Unfortunately parallel studies in hemophilic subjects by the antibody and biological assay technique have not yet been reported, so the discrepancy cannot be closely examined; but if there is a real discrepancy between the results of the two methods, it might be due to the presence in normal plasma of material antigenically indistinguishable from factor VIII, yet lacking its biological activity. This might be a precursor, with a longer biological survival in the plasma. At least both these lines of work (A2, G1) are compatible witahthe hypothesis of a CRM-positive but biologically inactive form of factor VIII in normal serum (Gl), and perhaps in normal plasma (A2). The findings of Goudemand et al. (Gl) further suggest that classical hemophilia is CRM-positive but that von Willebrand’s disease is CRMnegative. This finding would seem to support Graham’s “regulatory” hypothesis (Fig. 2) rather than his “combining subunit” model (Fig. 3), for the regulatory hypothesis suggests that in von Willebrand’s disease the operon on the X-chromosome may be completely suppressed.
204
G. I. C. INGFUM
McLester and Wagner (Mlb) have since reported that the plasma from several hemophilic dogs was CRM-negative when tested with a rabbit antiserum prepared with pursed canine factor VIII. They also suggest that CRM-positive reactions of the type reported by Goudemand et al. (Gl) might be due to the previous immunization of the hemophiliacs with isoantigens other than factor VIII, in plasmas with which they had previously been infused for the treatment of bleeding episodes (the patients St. and De. had in fact both received numerous transfusions). This appears to be an important criticism, but to explain the findings of Table 2 it would be necessary to suppose that the patients with von Willebrand’s disease had lacked some isoantigen other than “antigenic VIII” present in the hemophiliacs and also in the normal subjects, to which De. had become immunized. Barrow and Graham (Bl) feel that the experimental evidence tends to favor the combining subunit model, and that the failure to identify other genetic systems in man like their regulatory model makes this latter hypothesis rather less likely. For what the immunological evidence is worth it appears to favor the regulatory model, but more data are clearly required. A model analogous to Graham’s combining subunits has been proposed for hemoglobin synthesis (Rl). 2.5. THEFACTOR VIII Locus
ON THE
X-CHROMOSOME
Information available about the human X-chromosome was reviewed by Stewart a t the end of 1962 (52). The recognition of the X-linked blood group system Xg (M2) had stimulated a number of workers to investigate the Xg group in families transmitting abnormalities due to mutations of X-chromosome genes. A large combined study (Dl) covered 35 hemophilic families in which the bloods of 245 members were tested with anti-Xg, and useful information about linkage was obtained from 10 families. On the basis of this and other work, the order of genes which can be topographically related to hemophilia, with estimates of the distances in centimorgans which separate them, has been given (Kl) as Xg-27-G.6P.D’-5-deutan color vision-12-hemophilia. The gene for X-linked ichthyosis has been estimated as lying 23 centimorgans from Xg (Kl). A tentative map of the X-chromosome has recently been published by Ledinot and FrBzal (L2), who suggest that the Xg and deutan loci are in the short arm and that the loci for G.GP.D, protan color vision, and factors VIII and IX are all in the long arm of the chromosome, in that order, but the order quoted above is probably better founded (Ml). 1
G.6P.D = glucose-6-phosphate-dehydrogenasedeficiency.
BLOOD-COAGULATION FACTOR VIII
205
REFERENCES, SECTION 2 A l . Abildgaard, C. F., Cornet, J. A., Fort, E., and Schulman, I., The in vivo longevity of antihaemophilic factor (factor VIII). Brit. J. Haematol. 10, 225-237 (1964). A2. Adelson, E., Rheingold, J. J., Parker, O., Steiner, M., and Kirby, J. C., The survival of factor VIII (antihemophilic globulin) and factor IX (plasma thromboplastin component) in normal humans. J. Clin. Invest. 42, 1040-1047 (1963). A3. Aggeler, P. M., White, S. G., Glendening, M. B., Page, E. W., Leake, T. B., and Bates, G., Plasma thromboplastin component (PTC) deficiency: a new disease resembling hemophilia. PTOC.SOC.Exptl. Bwl. Med. 79, 692-694 (1952). A4. Alexander, B., and Goldstein, R., Dual hemostatic defect in pseudo-hemophilia. J. Clin. Invest. 82, 551 (1953). B1. Barrow, E. M., and Graham, J. B., Von Willebrand’s disease. Progr. Hematol. 4, 203-221 (1964). B2. Biggs, R., and Matthews, J. M., The treatment of haemorrhage in von Willebrand’s disease and the blood level of factor V I I I (AHG). Brit. J. Haematol. 9, 203-214 (1963). B3. Biggs, R., Douglas, A. S., Macfarlane, R. G., Dacie, J. V., Pitney, W. R., Merskey, C., and O’Brien, J. R., Christmas disease: a condition previously mistaken for haemophilia. Brit. Med. J. 11, 1378-1382 (1952). B4. Blackburn, E. K., Primary capillary haemorrhage (including von Willebrand’s disease). Brit. J. Haematol. 7 , 239-249 (1961). B5. Brinkhous, K. M., Chairman’s opening remarks. I n “Hemophilia and Hemophilioid Diseases” (K. M. Brinkhous, ed.), pp. xxi-xxii. Univ. of North Carolina Press, Chapel Hill, North Carolina, 1957. B6. Brinkhous, K. M., and Graham, J. B., Hemophilia in the female dog. Science 111, 723-724 (1950). B7. Bulloch, W., and Fildes, P., Hemophilia. In “Treasury of Human Inheritance” (Eugenics Laboratory Memoirs XII). Parts V-VI, Dulau & Co., London, 1911. B8. Barrow, E. M., Heindel, C. C., Roberts, H. R., and Graham, J. B., Heterosygotes and homosygotes in von Willebrand’s disease. PTOC.SOC.Exptl. Bwl. Med. 119, 79-82 (1965). C1. Cornu, P., Larrieu, M. J., Caen, J., and Bernard, J., Maladie de Willebrand. &ude clinique, g6n6tique et biologique (& propos de 22 observations). Nouvelle Rev. Franc. Hematol. 1, 231-262 (1961). C2. Cornu, P., Larrieu, M. J., Caen, J., and Bernard, J., Transfusion studies in von Willebrand’s disease: effect on bleeding time and factor VIII. Brit. J . Haematol. 9, 189-202 (1963). D1. Davies, S. H., Gavin, J., Goldsmith, K. L. G., Graham, J. B., Hamper, J., Hardisty, R. M., Harris, J. B., Holman, C. A., Ingram, G. I. C., Jones, T. G., McAfee, L. A., McKusick, V. A., O’Brien, J. R., Race, R. R., Sanger, R., and Tippett, P., The linkage relations of hemophilia A and hemophilia B (Christmas disease) to the Xg blood group system. Am. J. Human Genet. 16, 481-492 (1963). D2. Dreyfus, J. C., and Schapira, G., Les m6canisma de la biosynthbse des prot6ines. Path. Bwl. 13, 395402 (1965). El. Egeberg, O., Changes in the activity of antihemophilic A factor (f. VIII) and in the bleeding time associated with muscular exercise and adrenaline infusion. Scand. J. Clin. Lab. Invest. 16, 539-549 (1963). E2. Epstein, C. J., Structural and control gene defects in hereditary diseases in man. Lancet 11, 1066-1067 (1964).
206
(3.
I. C. INQRAM
F1. Field, R. A., Rickard, C. G., and Hutt, F. B., Hemophilia in a family of dogs. Cornell Vet. 86, 285-300 (1946). GI. Goudemand, M., Foucaut, M., Hutin, A., and Parquet-Gernez, A,, Les anticoagulants circulants anti-facteur VIII au coum de l’hemophilie A (nature et mecanisme d’action.). Wouvelle Rev. Franc. Hematol. 3, 703-722 (1963). G2. Graham, J. B., Biochemical genetic speculations provoked by considering the enigma of von Willebrand’s disease. Thromb. Diath. Haemorrhag. 9, Suppl. 2, 119125 (1963). G3. Graham, J. B., McLendon, W. W., and Brinkhous, K. M., Mild hemophilia: an allelic form of the disease. Am. J. Med. Sci. 226, 46-53 (1953). G4. Graham, J. B., McLester, W. D., Pons, K., Roberts, H. R., and Barrow, E. M., Genetics of vascular hemophilia and bioaynthesis of the plasma antihemophilic factor. I n “The Hemophilias,” Intern. Symp., Washington, D.C., 1963 (K. M. Brinkhous, ed.), pp. 263-275. Univ. of North Carolina Press, Chapel Hill, North Carolina, 1964. HI. Hellem, A., and Owren, P. A.,B The mechanism of the hemostatic function of blood platelets. Acta Haematol. 81, 230-238 (1964). H2. Hutt, F. B., Rickard, C. G., and Field, R. A., Sex-linked hemophilia in dogs. J . Hered. 89, 3-9 (1948). Jl. Jones, J. H., Rizza, C. R., Hardisty, R. M., Dormandy, K. M., and Macpherson, J. C., Combined deficiency of factor V and factor VIII (antihaemophilic globulin). A report of three cases. Brit. J . Haemabl. 8, 120-128 (1962). 52. Jurgens, R., Lehmann, W., Wegelius, O., Eriksson, A. W., and Hiepler, E., Mitteilung uber den Mange1 an antihiimophilem Globulin (Faktor VIII) bei der Aaliindischen Thrombopathie (v. Willebrand-Jiirgens). Thromb. Diath. Haemorrhag. 1, 257-260 (1957). K1. Kerr, C. B., Wells, R. S., and Sanger, R., X-linked ichthyosis and the Xg groups. Lancet 11, 1369-1370 (1964). L1. Larrieu, M. J., and Soulier, J. P., D6ficit en facteur anti-h6mophilique A chez une ale, associ6 8, un trouble du saignement. Rev. Hematol. (Paris) 8, 361-370 (1953). L2. Ledinot, E., and Frezal, J., fibauche de carte factorielle du chromosome X. Ann. Genet., Paris 7 , G.lO1-105 (1964). L3. Lyon, M. F., Gene action in the X-chromosome of the mouse (Mus musculus, L.) Nature 190, 372-373 (1961). M1. McKusick, V. A., “Human Genetics,” p. 52. Prentice-Hall, Englewood Cliffs, New Jersey, 1964. Mla. McLester, W. D., and Graham, J. B., Gene expression in heterozygotes and synthesis of plasma antihaemophilic factor. Nature 201, 1040-1042 (1964). Mlb. McLester, W. D., and Wagner, R. H., Antibody to antihemophilic factor and its lack of reaction with hemophilic plasma. Am. J. Physwl. 208, 499-507 (1965). M2. Mann, J. D., Cahan, A., Gelb, A. G., Fisher, N., Hamper, J., Tippett, P., Sanger, R., and Race, R. R., A sex-linked blood group. Lancet I, 8-10 (1962). M3. Miller, S. P., and Siggerud, J., Abnormal blood coagulation in carriers of hemophilia. J. Lab. Clin. Med. 68, 621-637 (1964). M4. Muhrer, M. E., Lechler, E., Cornell, C. N., and Kirkland, J. L., Antihemophilic factor levels in bleeder swine following infusions of plasma and serum. Am. J. Physiol. 208, 508-510 (1965). Nl. Nilsson, I. M., Blombiick, M., and Blomback, B., v. Willebrand’s disease in Sweden. Its pathogenesis and treatment. A d a Med. Scand. 164, 263-278 (1959). N2. Nilsson, I. M., Blombiick, M., and von Francken, I., On an inherited autosomal
BLOOD-COAGULATION FACTOR VIII
207
hemorrhagic diathesis with antihemophilic globulin (AHG) deficiency and prolonged bleeding time. Aeta Med. Scand. 169, 35-57 (1957). N3. Nilsson, I. M., Blomback, M., Jorpes, E., Blomback, B., and Johansson, S.-A., v. Willebrand’s disease and its correction with human plasma fraction 1-0. Acla Med. Scand. 169, 179-188 (1957). 0 1 . Otto, J. C., An account of an hemorrhagic disposition existing in certain families. Med. Reposil. (N.Y., 2nd ed.) 6, 1 4 (1803). P1. Pitney, W. R., Kirk, R. L., Arnold, B. J., and Stenhouse, N. S., Plasma antihaemophilic factor (factor VIII) concentrations in normal families. Brit. J . Haematol. 8, 421-428 (1962). P2. Preston, A. E., and Barr, A., The plasma concentration of factor VIII in the normal population. 11. The effects of age, sex and blood group. Brit. J . Haematol. 10, 238-245 (1964). P3. Parks, B. J., Brinkhous, K. M., Harris, P. F., and Penick, G. D., Laboratory detection of female carriers of canine hemophilia. Thromb. Diath. Haemorrhag. 12, 368375 (1964). R1. Reider, R. F., and Naughton, M. A., Hemoglobin G ( ~ ~ i t i ~ , , a~ enew ): abnormal hemoglobin, and an additional individual with four hemoglobins, Bull. Johns Hopkim H08p. 116, 17-32 (1965). S1. Skilhegg, B. A., Hellem, A. J., and Odegaard, A. E., Investigations on adenosine diphosphate (ADP) induced platelet adhesiveness in vitro. Part 11. Studies on the mechanism. Thromb. Diath. Haemorrhag. 11, 305-316 (1964). 52. Stewart, J. S. S., The X chromosome of man. Lancet 11, 1269-1272 (1962). V1. von Willebrand, E. A., Hereditare Pseudohamofili. Finska LCikaresdUskap. Handl. 68, 87-112 (1926). V2. von Willebrand, E. A., Vber hereditare Pseudohamophilie. Acla Med. Scand. 76, 521-550 (1931). V3. von Wdebrand, E. A., and Jurgens, R., Vber ein neues vererbbares Blutungsiibel: die konstitutionelle Thrombopathie. Deut. Arch. Klin. Med. 176, 453483 (1933).
3. Physiological Control
3.1. SYNOPSIS Although little is known of the regulatory mechanisms maintaining the plasma concentrations of the clotting factors at their normal levels in a given individual, various circumstances have been recorded in which their levels are observed to rise. For factor VIII, some involve relatively long-term changes, such as pregnancy, inflammatory states, and hyperthyroidism, while very sudden but short-lived increases follow hard muscular exercise or the injection of adrenaline. These changes are now being extensively investigated, and attempts have been made to elucidate their mechanism. 3.2. LONG-TERM CHANGES Under this heading it is possible to do little more as yet than record the situations in which a maintained rise in plasma factor VIII concentration has been observed. It seems worthwhile to do this because these facts will
208
G . I. C. INGRAM
form part of the corpus of data for which future hypotheses must be able to account and against which they will be tested.
3.2.1. Pregnancy Factor VIII levels up to 4 times normal have been reported at term (P5, 53, T3). Some other clotting factors also rise in concentration during pregnancy, for instance fibrinogen (16, M3, P5), factor VII (Ll, Nl), and factor X (Pl), but apparently not prothrombin (Dl). The significance of this pattern of changes is not known. 3.2.2.Chronic Inflammatory States 0. Egeberg has estimated factor VIII concentrations in a number of clinical conditions. Among those that may be loosely grouped together as “chronic inflammatory,” he has found an increase in factor VIII concentration in patients with atherosclerotic heart disease (E5), diabetes (E15), and miscellaneous chronic disorders (E3).In each study concomitant increases in factor VIII and fibrinogen were found, and in the second and third there was also some increase in factor V among the patients studied. An increase in the bleeding-time factor has been described in diabetes (02), in atherosclerotic disease (03)and in multiple myeloma (P6). 3.2.3.Hyperthyroidism Egeberg (E9, E17) found that the plasma concentration of factor VIII was correlated with the basal metabolic rate (BMR)over the range of values obtained from 4 hypothyroid and 11 hyperthyroid patients, lying between 2 and 4 times the average normal concentration in the latter group. The rise in concentration closely paralleled the rise in B M R when hypothyroid patients were treated with thyroxine. It was most interesting that the bleeding time was prolonged in hypothyroid patients, and fell to normal on thyroxine treatment, in view of the association of a long bleeding time with a low factor VIII concentration in von Willebrand’s disease (see Section 2.2.1,above), and the opposite association in atherosclerosis and diabetes. 3.2.4. Oral Anticoagulants A moderate rise in factor VIII concentration occurs in patients treated with the oral anticoagulants, which may continue until slightly after the end of treatment (E4, P3). 3.2.5. Dietary Changes in Animals
Davidson et al. (D2, D3) fed an atherogenic or a thrombogenic diet to groups of rats for periods of a month or more, in each case with and without
BLOOD-COAGULATION FACTOR VIII
209
added phenindione (2-phenylindane-l13-dione). In all four groups the mean factor VIII concentration rose by comparison with the rats on a control diet: with the atherogenic diet, containing arachis oil, the concentration rose about 2-fold, and with the thrombogenic diet, containing butter fat, the increase was 34-fold; the presence of phenindione did not affect the factor VIII response. Similar results were obtained by Merskey and Wohl (M5) with a diet enriched with beef fat. 3.3. SHORT-TERM CHANGES 3.3.1. E$ect of Adrenaline In 1903, Vosburgh and Richards were studying the effect of adrenaline injections on the blood sugar concentration in anesthetized animals, and noted that blood samples withdrawn from animals which had received adrenaline tended to clot in the syringe before they could be ejected into test tubes. After further investigation, they reported (V2) a definite shortening of the clotting time following adrenaline administration. Ten years later, their work stimulated W. B. Cannon to undertake further studies in anesthetized or decerebrate cats, hares, and rabbits, for which he and Mendenhall (C2) constructed a special coagulometer to measure and record the clotting time of arterial blood samples. In a series of eight papers over 3 years, Cannon and his associates reported most detailed investigations into the effects on blood clotting of adrenaline injection and of various procedures calculated to liberate adrenaline in the body (Cl, C3, C4, D4, G3, G4, M4, 52). Although their work was left unfinished, they established that the intravenous injection of adrenaline a t loF6g/kg, as well as hemorrhage and splanchnic stimulation, led to a shortening of the clotting time of whole blood. During the next 40 years this work was referred to from time to time and, as new clotting tests were developed, various workers attempted to elucidate the mechanism of this response to adrenaline (F3, M6, P2, R2, T1, T2, U1, W2) but with rather inconclusive results; this and other work has been summarized by Fonvell (F2). In 1958, Christie et al. (C5) made a series of daily assays by the Biggs method (Bl) of factor VIII concentrations in the blood of a hemophiliac being treated with animal antihemophilic material during the period of wound healing following abdominoperineal resection of the rectum for adenocarcinoma. The hemophilia was of only moderate severity, and the usual factor VIII concentration was about 10% of average normal. At the operation there had been some diaculty in securing a blood vessel high in the perineal wound, and after considerable blood loss the wound had been packed. The only bleeding during the postoperative period occurred on two occasions when the packing was removed, and then sudden, serious hemor-
210
G . I. C. INGRAM
rhage was provoked. Christie et al. noted that on the day of operation and on these two other days the responses to the antihemophilic treatment were unusually high. It seemed possible that either the bleeding itself, or the blood transfusion it had necessitated, had somehow induced an increase in the patient’s own factor VIII, since it became clear on the subsequent days of treatment that the relatively small variations in the amounts of antihemophilic material given could not have been responsible for the differences in the responses observed. It was thought that if hemorrhage had been the stimulus responsible, its effect might have been mediated through adrenaline release (G5) and the Cannon effect. Accordingly, another hemophiliac of comparable severity, subject Bat. of Ingram (I4), received an intravenous infusion of adrenaline over 15 minutes, after which the factor VIII concentration was found to have risen from 14% to 40%. Studies were then made on other hemophiliacs of varied severity, and on g/min normal subjects, infusing adrenaline a t the rate of 10-16 X for 10-15 minutes. Three severely affected hemophiliacs, with no detectable factor VIII in the plasma, showed none after adrenaline; but, of five other less severely affected patients, four showed a rise as had subject Bat., and a comparable result was obtained in normal males, the mean rise in 13 subjects being to 176% of the initial level (14). The effect appeared to last for a few hours. These results were confirmed by Egeberg (E16). 3.3.2. E$ect of Exercise Kesseler and Egli (Kl) had shown that the clotting time of whole blood was shortened following muscular exercise. At about the time when the above studies were being made with adrenaline, but quite independently, Rizza (Rl) extended this finding by noting that samples of blood taken from normal subjects following severe exercise yielded higher factor VIII concentrations on assay than samples taken from other subjects. He found that a relatively short period of brisk muscular exercise roughly doubled the plasma factor VIII concentration, and thus obtained a very similar effect to that following adrenaline infusion, both in normal subjects and in mildly affected hemophiliacs. This work was confirmed by Iatridis and Ferguson (11) and by Egeberg (E10) using a different assay system; Egeberg observed immediate increases in factor VIII concentration of 2-3.5-fold following strenuous exercise of only 3-minute duration. The effect began to pass offimmediately and was completely lost by the following day. Parallel assays or determinations were made for fibrinogen, prothrombin, and factors V, VII, IX, XI, and XII, but no comparable change was observed in any. It was particularly interesting that the rise in factor VIII was not here accompanied by a corresponding rise in fibrinogen concentra-
BLOOD-COAGULATION FACTOR VIII
211
tion, as had been observed under other circumstances (see Section 3.2, above). Egeberg also demonstrated a rise in factor VIII concentration in patients with von Willebrand’s disease, together with a shortening of the bleeding time (E14), and in one patient with thrombasthenia (ElS), who were exercised in a similar manner. 3.3.3. Other Acute Slates Egeberg studied a number of acute states and found that a rise of factor VIII was associated with the induction of nonspecific fever (E6, ElO), intravenous infusion of serum (E7), intramuscular injection of blood (E8) or intravenous infusion of hemolyzed blood (E2), drug-induced diuresis in a patient with cardiac edema ( E l l ) , and surgical operations (El). In the very acute changes, factor VIII alone rose in concentration, but where the process occupied several days there were normally associated increases in fibrinogen and in factor V. Other clotting factors remained at their previous levels. These investigations were all made in persons with normal hemostasis. THE NATURE OF THE ACUTERISE IN FACTOR VIII CONCENTRATION Experimental study of the rise in factor VIII concentrations, now known
INTO 3.4. INVESTIGATIONS
to occur in so many situations, has been almost entirely confined for practical reasons to a study of the effects of exercise or adrenaline administration. These situations, it will be recalled, are those in which a n isolated rise in factor VIII can be observed, unaccompanied by changes in fibrinogen or factor V. It is thus uncertain to what extent the results may be applied to the more protracted, combined increase in fibrinogen and factors V and
VIII. First, it should be remembered that the basic observation, a rise in plasma factor VIII concentration, depends entirely upon a biological assay, in which the corrective effect of the test plasma is compared to that of a control plasma when both are added to plasma obtained from a severely affected hemophiliac, or to an artificial system containing necessary clotting factors other than factor VIII. I n acute experiments it has been usual to assay the subject’s pretreatment plasma as well as the plasma obtained after the experiment, or to use the pretreatment plasma as the “standard” (nominally 100%) for the assay. The second procedure eliminates errors due to differences between subjects, but uncertainty still remains regarding the effects of the experimental treatment upon the assay system, apart from a possible true increase in factor VIII concentration. A number of experiments have therefore been directed to testing the validity of the
212
G . I. C. INGRAM
assay system under the conditions of experiment-i.e., when plasma from exercised or adrenaline-treated subjects is added to it. In other work, the mode of action of the stimulus has been studied in the tested subjects; other tests have been carried out on the plasma obtained from them. A special investigation was made of a local increase in factor VIII concentration in congested veins. 3.4.1. Validity of the Assay System under These Experimental Conditions
The most obvious source of biological invalidity in the assay would be a direct effect upon the system of either adrenaline itself or some component other than factor VIII, of which the blood concentration rose as a result of adrenaline treatment or exercise. These possibilities have been fairly fully investigated and it has not proved possible to explain the effects in this way. Adrenaline itself has been found to accelerate blood coagulation in vitro (F3, 14, W2), but only a t concentrations of an order of magnitudeone dex (A3)-higher than those produced in vivo. Noradrenaline, glucose, fructose, insulin, histamine, and lactate were found to have no effect (F3, 14). An especially important component to investigate has been plasmin, the proteolytic enzyme known to appear in the blood after adrenaline administration or exercise (B2, K2). The plasmin response passes off much more quickly than the factor VIII response (Rl, V1) and, even in the initial phase, shows no correlation in magnitude with the rise in factor VIII (I4),so it seems clear that an explanation in terms of plasmin may be discounted. Another possibility is that some other clotting factor is increased or activated in such a way that the assay system responds fortuitously to it in a way indistinguishable from the usual response to factor VIII, e.g., factors XI and XII; factor XI1 is known to rise on exercise (11). That this might occur over a limited range of the dose-response curve in the thromboplastin generation test system was shown by experiments in which the addition of “activation product” (Wl) simulated an increased factor VIII concentration (author’s unpublished observations, 1960; Fl), although statistical invalidity would probably be detectable over a series of experiments if this were the explanation. This also was looked for, but was not found (14). It is interesting that, in a patient with severe factor VIII deficiency and partial factor X I deficiency (Sl), adrenaline infusion was followed by a marked rise in factor XI concentration and the appearance of a trace of factor VIII (K. Schulz, personal communication, 1964). Furthermore, the confusion that arose some years ago over factor IX assay now seems to have been due to activation of the contact factors (P4), hence
BLOOD-COAGULATION FACTOR VIII
213
there is good reason to investigate the same possibility in the assay of factor VIII. However, the exercise effect has been demonstrated with assays involving (supposedly maximal) activation of factors X I and XI1 with kaolin (I3), and a system to which activation product was added (Vl) duly registered the adrenaline effect. Furthermore, a rise in factor VIII concentration has now been produced by exercise in patients with severe deficiencies of factors X I and XII. Egeberg (E13) tested two patients with gross deficienciesof factor XI, and Goudemand et al. (G2) and A. Parquet-Gernez (personal communication, 1964) tested factor XII-deficient patients. Finally, it might be that some component not recognized or even detectable as a clotting factor, yet able to reinforce factor VIII concentration in the assay, might be released by adrenaline. If this occurred, it would be reasonable to expect it to be unconnected with hemophilia, and it might therefore be released as well in hemophiliacs as in normal subjects, although of course in severely affected patients it would not normally be detectable. To investigate this possibility, two severely affected hemophiliacs were treated with adrenaline, blood samples being taken before and afterward. The treated and untreated hemophilic plasmas were then used to make a series of dilutions of normal untreated plasma, and the dilutions in the treated hemophiliac plasma assayed against those in the untreated one, but no difference in factor VIII concentration was found (the results in one hemophiliac subject have been reported (14); the experiment has since been repeated in another hemophiliac). Another experiment (14), repeated twice in the same hemophiliac, was to precede the adrenaline infusion with an infusion of normal plasma; in neither instance did the adrenaline treatment further increase the factor VIII concentration in the subject’s blood. Thus none of these experiments has produced evidence that the assayed rise in factor VIII can be discounted as spurious. Nevertheless this possibility should not be forgotten, for it may simply be that the correct experiments to demonstrate this have not yet been contrived. The ultimate test of an increase in factor VIII activity is the demonstration of greater hemostatic value in hemophilic bleeding; this is unfortunately a very difficult test to apply. 3.4.2. Other Observations on Tested Subjects A number of interesting observations have been carried out in tested subjects. Egeberg (E13, E16) repeated the exercise test on 6 successive days in three normal subjects, and showed that successively lower responses were obtained, with a high response again after a subsequent interval of
214
G. I. C. INGRAM
4 days without exercise. Similarly, six normal subjects were exercised on 5 successive days and treated with adrenaline on the sixth day; the average adrenaline response was considerably lower than would otherwise have been expected and also lower than the initial response to exercise (E16). Egeberg was able, however, to obtain a marked response to exercise in two normal subjects who were already showing a moderate response to artificially induced fever, although the total response was not more than might have been obtained by a single stimulus alone (E10). These resulk suggest that the mechanism, whatever it is, can be exhausted fairly readily. Goudemand et al. (Gl) considered the possibility that a quantity of factor VIII might be released from the spleen in response to adrenaline administration or release. Therefore they tested a series of previously splenectomised subjects; in seven of them a rise in factor VIII after exercise could not be demonstrated, but adrenaline infusion produced a mean 2-fold increase in 11. Vaughan Jones et al. (Vl) studied the effect of adrenaline-blocking drugs. Ahlquist (A2) had suggested that the action of adrenaline might be divided into two groups, the a-effects which include vasoconstriction and other excitatory functions, and the @-effectsof vasodilation and cardiac stimulation. The a-effects may be blocked by phentolamine [2-N-(3-hydroxyphenyl)-p-toluidinomethyl-2-imidasoline]and the @-effectsby pronethalol [2-isopropylamino-l-(2-naphthyl)ethanol].Vaughan Jones et al. (Vl) gave adrenaline infusions to a small series of subjects, with and without these blocking drugs. The factor VIII response appeared to be blocked by pronethalol but not by phentolamine, which suggests that it was a /3-effect. 3.4.3. Tests on Plasma from Treated Subjects Various experiments have been concerned with transfusing plasma from exercised donors into factor VIII-deficient subjects. Risza (Rl) gave plasma from exercised donors to two hemophiliacs, and reported a more prolonged survival than would be expected after an ordinary plasma infusion; Egeberg (E14) noted a similar phenomenon when transfusing a patient with moderately severe von Willebrand's disease. Egeberg also gave whole blood from an exercised donor to a similar patient, and reported a more pronounced and longer-lasting effect on the bleeding time than had previously been obtained with ordinary transfusions, or by exercising the patient. He also showed that this effect resided in the plasma and not the platelets of blood from exercised donors, and that it could also be demonstrated in serum (E16). In other, more severely affected von Willebrand patients, however, the bleeding-time response was not oht,ained, and Egeberg
BLOOD-COAGULATION FACTOR VIII
215
raises the possibility that these may have been homozygotes; in one for whom the data are given, it seems that the patient did not synthesize factor VIII following the infusion (see Section 2.2.3, above). These observations suggest that plasma does not itself contain the bleeding-time factor, but stimulates at least the von Willebrand heterozygote to form it, and that this ability is heightened by exercise. It would now be possible to test this hypothesis in vitro with critical concentrations of adenosine diphosphate (ADP) (01,02). However this may be, all the foregoing observations suggest that the increased biological activities recorded as raised concentrations of factor VIII and bleeding-time factor in exercised normal subj ects and von Willebrand patients, respectively, can still be recognized after plasma from a treated subject is transfused into factor VIII-deficient subjects. “Passage” through another subject in this way might be expected to remove extraneous material, so that what is “recovered” as factor VIII activity after passage is more credibly acceptable as such. The same argument may be applied rather more forcefully to the bleeding-time factor, for here the test (the bleeding time) is a genuine measure of hemostatic competence. Other evidence, however, makes this conclusion more doubtful. Egeberg (E10, E16) dialyzed plasma samples from exercised persons, and showed that the high factor VIII concentration survived this treatment without loss. A quantity of plasma taken after exercise was then divided into two portions, one of which was dialyzed and the other maintained under similar conditions undialyzed. The factor VIII concentrations in the subsamples remained closely similar, but, on transfusing them into a patient with von Willebrand’s disease, the dialyzed sample (given first) produced no change in the patient’s factor VIII concentration or bleeding time, whereas the undialyzed sample, given 2 hours later, was rapidly followed by a rise in factor VIII and a shortening of bleeding time. This important experiment clearly requires confirmation, because it casts serious doubt on the assumption that biological assay dependably identifies “true” factor VIII activity in these experimental conditions. Independent work calling the same assumption in question has been reported by Goudemand et al. (G2). It will be recalled that these authors employed a naturally occurring anti-VIII antibody to detect material in plasma and serum having the antigenic characteristics of factor VIII even if lacking its biological activity (Section 2.4, above). This group (Gl) have since applied their test to the plasma of 11 normal subjects before and after exercise. While ordinary bioassays yielded a mean increase of 237% in factor VIII concentrations, neutralization of the antibody increased by only 117%, a figure the authors do not consider to represent a significant
216
G. I. C. INGRAM
rise. The extreme complexity of these experiments must be remembered, but on their face value the results suggest that the rise in biological activity did not indicate an absolute increase in factor VIII concentration. 3.4.4. Efect of Venous Stasis Egeberg (E12) applied a blood-pressure cuff inflated to 90 mm Hg for 10 minutes to one upper arm in normal subjects, taking blood samples from antecubital veins in both arms. Observations on the hematocrit and plasma protein concentrations suggested that a definite hemoconcentration had been produced, but with loss of some plasma protein presumably by diffusion from the vascular bed. Parallel assays of a number of clotting factors were undertaken in the paired samples. There was a general increase in concentration comparable to the effect on total plasma protein, but a much larger increase was observed uniquely (at least when silicone-coated glassware was used) in factor VIII. Egeberg suggested that his findings were due to tissue material entering the blood from the congested arm. Goudemand et al. (Gl) have also put forward this idea to explain the adrenaline response, believing that adrenaline causes tissue material to enter the circulation (54) and that this then activates a factor VIII precursor. 3.5. POSSIBLE MECHANISMS FOR THE RISE IN ASSAYABLE FACTOR VIII Readers who have followed the writer thus far will realize that this discussion now suffers from almost too much evidence. It is therefore important to try to set down the principal findings in order to summarize the present position and suggest the most useful lines for further observation. The facts of the long-term increase in factor VIII concentration are the less confused and, as before, will be taken first. 3.5.1. Long-Term Changes In pregnancy, diabetes, and “chronic inflammatory” states there is a sustained rise in factor VIII concentration, accompanied by increases in other clotting factors, of which fibrinogen is the most constant and probably the most closely correlated quantitatively. In diabetes and atherosclerotic disease there is also an increase in the bleeding-time factor. Similar effects on the clotting factors can be produced by applying “stresses” which operate over a few days, such as the intramuscular injection of the subject’s own blood. The rate of increase in factor VIII concentration indicated by serial assays is not more rapid than may be seen in patients with von Willebrand’s disease after infusion of hemophilic plasma, and may therefore be reasonably ascribed to an increased rate of synthesis, since there is
BLOOD-COAGULATION FACTOR VIII
217
no reason for the rate of removal from the plasma (“utilization”) to decrease. The determination of fibrinogen is relatively straightforward, and there is no reason to doubt that the plasma concentration of this protein rises about as rapidly as is postulated for factor VIII, or to question that this implies an increased rate of synthesis of fibrinogen.
3.5.2. Short-Term Chan,ges There are two special characteristics of the experimental short-term rise in factor VIII concentration, its extreme rapidity (e.g., following only 3 minutes of severe exercise) (E16), and that it is unaccompanied by an increase in any other hemostatic component except the bleeding-time factor. Hypotheses must therefore account for an apparent increase in plasma concentration up t o about 6-fold in a matter of minutes in this clotting factor alone. Could increased rate of synthesis account for an effect of this magnitude? McLester and Graham (M2), discussing their “regulatory” model for the genetic control of factor VIII synthesis (Section 2.3.2, above), suggested that a substance of small molecular weight like adrenaline might act as an inducer, which should normally inactivate the repressor of the autosomal operon (although the abnormal repressor formed in von Willebrand’s disease would not be inhibited); but even with this possible mechanism in view, the synthetic effort required to meet the experimental findings seems truly phenomenal and would require the support of strong direct evidence before serious acceptance. It also seems difficult to envisage a sufficiently rapid mechanism for flooding the circulation with factor VIII from some tissue reserve, and there is no apparent reason for so great a store to be maintained. NourEldin (N2) believed that the tissues contain much factor VIII, but, if his work is relied upon here, it is then necessary to explain why factor VIII does not appear in the blood of severely affected hemophiliacs after adrenaline infusion, for hemophilic tissue extracts will clot hemophilic blood as rapidly as extracts from normal tissue (B3), and so presumably contain the same clotting factor@). Goudemand et al. (Gl) inquire with reason, “Comment pourrait Ztre stock6 en permanence une quantit6 aussi 6norme de G.A.H. [i.e., factor VIII] telle que le taux dans le sang puisse &re multiplid par d h 6 fois%”2 The only other explanation in view a t present is that the acute response consists of a potentiation or activation of circulating factor VIII without true increase in quantity, and that bioassay systems fail to distinguish between the two postulated forms of the factor. There is evidence that “How could a reserve of factor VIII be maintained sufficient to provide for a 2- to &fold rise in blood concentration?”
218
0. I. C. INGRAM
factor VIII can be “activated” by thrombin (R3)and by activated factor IX (B4, L2, Ml), to give approximately the necessary increase in activity, but that activated factor VIII is unstable, whereas the activity generated by adrenaline or exercise survives incubation at 37°C for 30 minutes (14) or storage at -20°C for a month with two cycles of freezing and thawing (E10). The antibody experiments described in Section 3.4.3 (Al, Gl) support a potentiation hypothesis, but it requires to be supported also by the demonstration that factor VIII may be potentiated without loss of stability. The possibility of a tissue “activator” entering the circulation (E12, G1) requires further study. Egeberg’s (E16) experiment showing loss of effect in vivo after dialysis, despite unchanged in vitro activity, should be repeated, for this evidence suggests an alteration in the molecule involving a small (dialyzable) component. Evidence was given above (Section 3.4.1) that the exercise or adrenaline effect was demonstrated by assay systems in which precautions had been taken to provide full surface activation. The comparative data of both Ikkala (12, 13) and Ingram (15) show, however, that the effects were more marked when these precautions were not taken, and Ikkala (12) also showed that the effect was more marked when phospholipid was not added to the system, although it was still clearly seen in its presence. The reduction in the effect either by surface contact or by phospholipid, or both (I2), raises the suspicion that if even more efficient agents had been added to the assay the effect would have been eliminated altogether, but so far this has not been achieved. 3.5.3. Conclusion
It seems likely that the relatively slow rise in factor VIII concentration in response to subacute or chronic “stress” represents an increased synthesis, and is comparable in rate to the increased synthesis seen in von Willebrand’s disease after infusing normal or hemophilic plasma. The very rapid increase which can be produced by exercise or adrenaline infusion is more likely to be due to potentiation of factor VIII in the blood stream, but a type of potentiation satisfying the experimental data does not yet appear to have been demonstrated. REFERENCES, SECTION 3 A l . Adelson, E., Rheingold, J. J., Parker, O., Steiner, M., and Kirby, J. C., The survival of factor VIII (antihemophilic globulin) and factor IX (plasma thromboplastincomponent) in normal humans. J . Clin. Invest. 42, 1040-1047 (1963). A2. Ahlquist, R. P., A study of the adrenotropic receptors. Am. J . Physiol. 163,586-600 (1948). A3. Allen, C. W., Dex. Observatory 71, 157 (1951).
BLOOD-COAGULATION FACTOR VIII
219
Bl. Biggs, R., Assay of antihaemophilic globulin in treatment of haemophilic patients. Lancet 11, 311-314 (1957). B2. Biggs, R., Macfarlane, R. G., and Pilling, J., Observations on fibrinolysis. Experimental activity produced by exercise or adrenaline. Lancet I, 402-405 (1947). B3. Brown, A., Personal communication, 1952; quoted by Biggs, R., and Macfarlane, R. G., “Human Blood Coagulation and Its Disorders,” 3rd ed., p. 226. Blackwell, Oxford, 1962. B4. Biggs, R., Macfarlane, R. G., Denson, K. W. E., and Ash, B. J., Thrombin and the interaction of factors VIII and IX, Brit. J . Haematol. 11, 276-295 (1965). C1. Cannon, W. B., and Gray, H., Factors affecting the coagulation time of blood. 11. The hastening or retarding of coagulation by adrenaline injections. Am. J . Phyrwl. 34,232-242 (1914). C2. Cannon, W. B., and Mendenhall, W. L., Factors affecting the coagulation time of blood. I. The graphic method of recording coagulation used in these experimentP. Am. J . Physiol. 34, 225-231 (1914). C3. Cannon, W. B., and Mendenhall, W. L., Factors affecting the coagulation time of blood. 111. The hastening of coagulation by stimulating the splanchnic nerves. Am. J . Physwl. 34,243-250 (1914). C4. Cannon, W. B., and Mendenhall, W. L., Factors affecting the coagulation time of blood. IV. The hastening of coagulation in pain and emotional excitement. Am. J . Physiol. 34, 251-261 (1914). C5. Christie, T. H., Graham-Stewart, C. W., and Ingram, G. I. C., Abdomino-perineal resection of the rectum in a haemophilic. Throm. Diath. Haemorrhag. 4, 224-234 (1960). D1. Davidson, E., and Tomlin, S., The levels of the plasma coagulation factors after trauma and childbirth. J . Clin. Pathol. 16, 112-114 (1963). D2. Davidson, E., Howard, A. N., and Gresham, G. A,, The nature of the coagulation defect in rats fed diets which produce thrombosis or experimental atherosclerosis. Brit.J . Exptl. Pathol. 43, 166-171 (1962). D3. Davidson, E., Howard, A. N., and Gresham, G. A., The effect of phenindione in rats fed diets which produce thrombosis or experimental atherosclerosis. Brit. J . Exptl. Pathol. 43,418-423 (1962). D4. Drinker, I